📄 Research720

huggingface.copaper

Your Benchmark Is an Empirical Measure Over Difficulty

The paper treats a benchmark as an empirical measure over item difficulty and formalizes two failure modes: measure mismatch (benchmark difficulty distribution differs from the intended one) and coverage gaps (uncovered difficulty regions that hide localized training progress). It proposes two diagnostic metrics with theoretical guarantees and analyzes 25 popular benchmarks across 7 domains using estimated item difficulties. It also introduces BenchFill, a difficulty-targeted item rewriting algorithm that reduces local coverage gaps.

evals, research· Added 31 Oct 2026

simonwillison.netarticle

dbreunig/nesformer

nesformer is a project by Drew Breunig that bakes a working Nintendo Entertainment System 6502 CPU emulator into the weights of Qwen3.5-4B, so NES ROMs such as Super Mario Bros run (very slowly) inside the model. It follows Divyansh Agrawal's qwenputer. Simon Willison relays the Opus 5.5 prompts Breunig used, and notes that Opus found Christopher Pow's CPU conformance tests.

research, coding-agents, models· Published and added 30 Oct 2026

en.cryptonomist.charticle

Meta’s AI-distilled model detector catches 79% at a 0.1% false-positive rate

WM-Basin is a Meta method for detecting student LLMs distilled from watermarked teacher models, described in "Watermark Basins: Black-Box Detection of Distilled LLMs" (arXiv:2610.37266, 27 October 2026). It feeds a suspect model's API text into a clean probe model and checks gradient alignment with a key-specific watermark basin. It reached 79% true-positive rate at 0.1% false-positive rate across 1,200 students, versus 29% for the prior token-level detector. Clean-data fine-tuning and paraphrasing weaken it.

research, training· Published and added 30 Oct 2026

huggingface.copaper

WorkFactory: Turning Professional Intents into Executable Work for Scalable Agent Training

WorkFactory is a framework that builds coherent, executable professional tasks for training working agents. It plans an information support graph from professional intents, then uses probe executions to find and repair missing or inconsistent task information. The authors built 112,686 tasks across 612 occupations and fine-tuned Qwen models at 4B, 27B and 35B-A3B (WorkPilot). WorkPilot-35B-A3B rose from 879 to 1565 Elo on GDPval-AA v2.1.

training, agents, research· Added 30 Oct 2026

github.comrepo

Show HN: I Turned DOOM into a Transformer

Doomformer compiles integer neural networks into ZDoom maps that run in ViZDoom. Voodoo dolls do the arithmetic, and sector light levels and plane heights hold the values. Python reads every value back and compares it with a NumPy integer reference at tolerance 0. It runs MNIST (97.16 % on all 10,000 digits, every pass exact), small transformers, a learning perceptron and a SUBLEQ machine. It is about 23,000 times slower than NumPy on a small MLP.

research, inference· Added 30 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

world2vec: Mechanistic world models through latent dynamics learning

world2vec is a method for learning mechanistic world models from observations and actions alone, without reward information. Its latent dynamics reproduce the true system over long prediction horizons using at least one order of magnitude fewer parameters than DreamerV3, TD-MPC2 and LeWM. The authors report state-of-the-art control of continuous environments from pixels and accurate system identification across a benchmark of 71 attractors.

research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

TextSeal2: Distortion-Free LLM Watermarking with Copula Mixing and Memoryless Races

TextSeal2 is a distortion-free watermark for LLM output that keeps its signal under speculative decoding. It merges two or more keys into one virtual key coupled to fresh noise through AutoMark's Gaussian copula, and verifies drafts with an extra exponential race. On three LLMs it detects 8 to 22 points more 100-token answers than TextSeal at a 10^-3 false-positive rate.

research, inference· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

RADIOlogy: A Single-Pass Approach to Training-Free Zero-Shot Anomaly Detection

RADIOlogy is a training-free, single-pass zero-shot anomaly detection method that uses multi-head attention pooling in vision-language models. It scores anomalies by contrasting average similarity between vision features and sets of normal and abnormal prompts, with no windowed execution. RADIO agglomerative features drive a dense refinement pipeline for pixel-level segmentation, and the method is applied to medical images.

research, multimodal· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

VeriCUA: Verification as Agentic Exploration for Computer-Using Agents

VeriCUA is a platform-agnostic framework that verifies task completion by computer-using agents through agentic exploration. The verifier inspects context to build a task rubric, then picks GUI or code checks for unresolved requirements and combines findings under explicit evidence rules. It works online and offline across desktop, mobile and web, and agrees better with reference labels than the baselines, with the largest offline gains on hard cases.

agents, evals, research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Multi-View MoE: Letting Every Router Shape Expert Computation

Multi-View MoE is a sparse mixture-of-experts design that uses several independently initialized routers to select experts, then reuses their relative scores to mix learned latent views into each selected expert's input. Each expert still runs one forward pass, so active-expert count, FLOPs and all-to-all volume are unchanged. On a 24-layer 10B-A1B model trained on 708B tokens, it gains 1.18 points on a nine-task mean; Multi-Router alone gains 0.6.

research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

On the Fundamental Limits of Prefill-Based Decoding Length Prediction

This paper derives bounds on predicting an LLM's decode length from prefill-phase hidden states, a primitive for inference scheduling. It models decoding as a stopped stochastic process, giving per-prompt normality and an affine variance law σ=αμ+b. Predictors (MLP, EGTP, FiLM, soft-label, Mixture-of-Experts) converge near the bound, so system error dominates model error. It proposes fixing model, data and sampling in benchmarks and reporting relative-error metrics.

research, inference, evals· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Efficient Scalable Oversight via Rewinding

Rewindable interactive proofs (RIPs) are a new model of relativizing interactive proofs for scalable oversight, in which a weak verifier can rewind an AI prover to an earlier saved state. The paper builds an RIP for computations robust to changes in a small fraction of oracle answers. The honest prover runs in linear time, the verifier runs in nearly linear time in the input length after one-time preprocessing, and no cryptography is used. A preliminary MNIST experiment shows an RIP can be faster than a deployed proof system.

research, ai-safety· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

PhysCount: Do MLLMs Count Visual Appearances or Physical Entities?

PhysCount is a benchmark of 546 real-world images and 4,595 human-written questions in 651 target-specific groups. It tests whether multimodal LLMs can separate visible appearances (direct view, reflections, glass) from the physical entities behind them. Across 40 MLLMs, counting appearances is much easier than linking them to entities, and the best model answers every question correctly in only 13.36% of groups.

evals, multimodal, research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Glance: The Generator Is Its Own Reranker

Glance is a RAG method in which the generator also reranks. It adds a lightweight decision adapter and pointer head that score candidate passages from one shared prefill, so a single resident 9B backbone does generation and reranking. After 39 minutes of MS MARCO fine-tuning it reaches nDCG@10 of 0.741 on SciFact and 0.345 on NFCorpus, against 0.721 and 0.326 for bge-reranker-v2-m3. On 300 HotpotQA questions it improves answer F1 by 0.045 over cross-encoder-selected context. Evaluation covers two retrieval datasets, one QA dataset and one backbone family.

llm-apps, research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Retrieving from Your Own Mistakes: On-Policy Training of Embedding Models for Multi-Hop Retrieval-Augmented Generation

On-policy state training is a method for training chain embedding retrievers for multi-hop RAG on the states the encoder itself reaches, including those built on its own wrong retrievals. Each state is labeled with the earliest missing gold passage, other missing gold passages are masked, and updates are contrastive and mixed with teacher-forced states. It retrieves complete evidence more often than teacher-forced training and graph-based systems, and transfers to harder long-chain HotpotQA and 2WikiMultiHopQA variants the authors built.

research, llm-apps, training· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Agentic LLM reasoning in pediatric diabetes outcome prediction from multimodal medical records

CEDAR (Clinical Evidence-Driven Agent for Risk-prediction) is an evidence-citing, self-auditing LLM pipeline for predicting pediatric diabetes outcomes at 24 months from 18 months of records. On 2,489 youth with type 2 diabetes and 4,018 with type 1, pooled AUROC was 0.815 versus 0.656 to 0.700 for trained ML in T2D, and 0.756 versus 0.603 to 0.631 in T1D. Only 4 of 70,246 cited values mismatched the chart. Two pediatric endocrinologists matched its predictions in about nine of ten cases.

research, agents· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

The Athlete as a Scientist of One: Agentic Bayesian Experimental Design for Individualized Physiology from Consumer Wearables

EpiCoach is an agentic system that treats one cross-country mountain biker as an N-of-1 experiment. Claims live in an evidence-gated hypothesis registry, and each must make a probabilistic prediction before a session. Predictions come from a digital twin fitted to heart rate, GPS and altitude. In a replay over eighteen months the twin beat standard forecasters on heart rate but not segment time, and the agent loop matched a static model. In the first three live weeks, the ledger sealed 20 predictions and scored 14, and a safety officer vetoed the agent twice. The registry, benchmark and protocol are released.

agents, research, evals· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Don't Lose the Thread: Empowering Long-Horizon LLM Agents with Cognitive Resource Self-Allocation

CORAL (Cognitive Resource Self-Allocation) is an agent-callable working-memory toolset that lets an LLM agent checkpoint task progress and verified facts, then reset its cluttered context and resume from a clean state. A Multi-episode Agentic Reinforced Policy Optimization algorithm trains the checkpointing. With an 8B model it reaches 42.8% average accuracy on GAIA, uses 6-7x fewer input tokens than a ReAct agent, and adds 8.0 points to a proprietary agent on BrowseComp.

agents, research, training· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

A Unified Hessian-Guided Framework for Post-Training Quantization

This preprint proposes a unified post-training quantization framework for large language models. It derives a Hessian-guided approximation of the output change caused by joint weight and activation quantization, using a tractable Fisher approximation. Quantization ranges, linear transformations and weight rounding are optimized together under one output-sensitivity objective. The authors report gains over state-of-the-art transformation and rounding methods in channel-wise and block-wise settings.

research, inference· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Decompose, Estimate, Select: Coverage-Aware Context Construction for Long-Form RAG

Decompose, Estimate, Select is a coverage-aware retrieval pipeline for long-form RAG. It splits a query into subquestions, ranks candidates with a coverage-trained bi-encoder, predicts a document-by-subquestion coverage matrix with a cross-encoder, and picks documents by greedy maximum-marginal coverage. On TREC RAG 2024, α-nDCG@10 rises from 0.625 to 0.702 (gold-nugget bound 0.718). A grounded self-feedback loop raises V_strict by 0.086 over two passes.

llm-apps, research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Only Accepted Tokens Count: An Energy Accounting of Speculative Decoding Under Latency Targets

This paper measures the energy cost of speculative decoding in SGLang with CUDA graphs on H100 and H200 GPUs, at batch sizes up to 64 across 5 corpora. Savings depend on drafted-token acceptance, which is lower on conversation than code and lower on Romansh than English. Savings shrink as batch size grows, and block drafters can push verification into the compute-bound regime, where speculation raises energy per token.

inference, research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Benchmarking color judgment in multimodal LLMs with CIE-computable ground truth

A five-task benchmark tests multimodal LLM color judgment against CIE colorimetric ground truth: AWB pair judgment, skin-tone bias direction, CCT estimation, Delta E 2000 pair ordering, and color naming under changing illuminants. Eight endpoints were tested. On ColorChecker-24 stimuli, closed frontier models reach CCT MAE of 715-1056 K versus 1358-2133 K for open-weight models, and a non-ColorChecker control shows this gap is partly template-dependent.

evals, multimodal, research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Turing LLM Post-Training: Calibrated Distillation and Efficient Dynamic-MoE Execution

Turing LLM post-training combines calibrated multi-teacher on-policy distillation (C-MOPD) with an execution path for dynamically routed mixture-of-experts models. C-MOPD matches each teacher on a local Top-K support with teacher-wise loss normalization and clipping. On Turing-20B-A2B it scores 0.9988 on four-domain integration, beats Qwen3-8B on math and code (85.92 vs 80.33 mean), and cuts consolidation step time from 398.7 to 139.3 seconds in a PPU profile.

training, research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

M-PFT: Angular Margin for Calibrated Prompt-Tuning of Fine-Tuned Vision-Language Models

M-PFT is an angular-margin method for prompt tuning of vision-language models that fixes base-to-novel dual miscalibration (underconfidence on base classes, overconfidence on novel classes). It adds two losses: minimizing the maximum cosine similarity between tunable class embeddings, and minimizing angular distance to each zero-shot embedding. Tests on 9 prompt-tuning methods and 15 datasets show lower novel-class calibration error at comparable accuracy and negligible overhead.

research, multimodal, training· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Aligning Medical Vision Language Models at the Capability Boundary with Entropy Adaptive Rubric Grounding

MedCBR is a two-stage alignment framework for medical vision language models. Difficulty Weighted Boundary Preference Optimization (DBPO) trains on counterfactual preference pairs at the model's capability frontier. Entropy Gated Grounded Policy Optimization (EGPO) then applies distilled clinical rubrics according to prediction entropy. It reaches 61.2% average accuracy on six medical benchmarks, with the largest gains on differential diagnosis, and the two stages combine super-additively.

research, training, multimodal· Added 31 Oct 2026, published 29 Oct 2026

blog.googlearticle

How Gemini is helping students and teachers in the classroom

Google reports two classroom studies of Gemini. In a 10-week RCT with over 1,100 UK students (ages 12-15) on Eedi, Gemini-based AI tutoring matched human tutors on in-platform tasks and quizzes, with expert approval of responses rising from 76% to 95%. Gains did not appear on STAR standardized tests. A yearlong Italian study of 1,600 teachers found lessons co-designed with Gemini shifted toward collaborative work.

research, evals, ai-policy· Added 30 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

RetroChessBench: A Benchmark for Reconstructing and Proving Chess Histories

RetroChessBench is a 12-task benchmark for retrograde chess: reconstructing legal game histories, exact-length constructions, unique endings and exhaustive answer sets, with references, rubrics and a terminal-reward RL environment. GPT 6 Astra High fails to give the required final answer in 22 of 24 attempts without tools and 18 of 24 with code execution. Mean terminal RL rewards are 14.75% and 27.19%. The authors link the failures to scarce worked derivations in web training data.

evals, research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

Sparse Autoencoders as Implicit Diversification Regularisers for Cross-Sectional Cryptocurrency Returns

This arXiv preprint (October 2026) tests a sparse autoencoder as a factor model for daily cryptocurrency returns, with equities as a secondary test. SAE portfolios show consistently lower volatility than common benchmarks, linked to lower correlations among selected assets. They do not beat a simple contrarian strategy on risk-adjusted return, and factor identities change between training runs.

research· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

DICE: Decentralized Incentive-driven Federated Learning for Efficient & Scalable Customization of Pretrained Models in Web3

DICE is a blockchain-powered federated learning (BCFL) algorithm for customizing large pretrained models among untrusted Web3 participants. It combines dynamic progressive learning and model slice verification, so hardware-limited participants can train part of the model, with a tree-structured committee consensus protocol and an efficiency-driven incentive scheme to scale to more data holders. Experiments cover vision and language models.

research, training· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper

PCM: A Proof-Carrying Message Protocol for Verifiable Multi-Agent Communication

Proof-Carrying Messages (PCM) is a protocol where each claim an AI agent sends carries evidence: records from an append-only Merkle log the agent cannot write, with inclusion proofs, signed by an attesting signer. Receivers verify offline in about 1 ms and label each claim proven, unproven or failed. On 926 messages (712 lies in 16 classes), PCM accepts 10 lies (1.4%), all stale evidence; plain A2A accepts all 712.

agents, research· Added 31 Oct 2026, published 29 Oct 2026

divagr.comwriteup

Qwenputer: A functioning, deterministic RISC-V CPU inside Qwen3.5

Qwenputer is a stock Qwen3.5 4B with a 32-bit RISC-V CPU (RV32IM plus Zba and a bignum unit) compiled directly into added layers, with no training. Gated DeltaNet states act as exact RAM via gate 1, beta 1 and one-hot keys. It scores 97.5 to 100% on DeepMind arithmetic sets, and 100% on NUPA long-integer arithmetic. It also has an experimental in-model filesystem. The checkpoint is 18.8 GB and needs the RvChat driver.

research, models· Published and added 29 Oct 2026

arxiv.orgpaper

TrussTutor-Bench: Benchmarking large language models on solution accuracy, sycophancy, and error detection in statics

TrussTutor-Bench is a benchmark of 200 verified two-dimensional truss problems from statics, paired with automatically generated student solutions. It evaluates thirteen LLMs on solution accuracy (64.5% to 97.0%), error detection (90.4% to 100%) and sycophancy. Models rejected their own correct solutions in 1.6% to 50.0% of cases when a student insisted the work was wrong. Sycophancy separated models; error detection did not.

evals, research· Added 31 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Gold vs.~Translation: Are We Measuring Language, Knowledge, or Locality in Translated Benchmarks?

Gold vs. Translation is an October 2026 arXiv study of whether DeepL-translated MMLU measures the same thing as native benchmarks. It compares seven pairings against six native benchmarks in five Nordic and Baltic languages, using 29 models. Translated MMLU adds predictive value for curriculum-overlap benchmarks such as Skolprov (ΔR²=.128) but not for Sweden-specific SwedishFacts (p=.242).

evals, research· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Leaving, Staying, or Living There? Duration Commonsense Reasoning with Human-LLM Co-annotations

This paper studies whether vision-language models can infer how long a person stays somewhere (living, staying, or leaving) from geotagged social media posts. The authors build a new corpus of geotagged tweets co-annotated by humans and LLMs. State-of-the-art models struggle even with multi-turn Chain-of-Thought prompting, often missing contextual cues and subtle hashtags while over-relying on prominent visual details.

research, evals, multimodal· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

inkGround: Benchmarking Fine-grained Visual Grounding Capabilities of Vision-Language Models for Cursive Handwriting in the Wild

inkGround is a benchmark for stroke trajectory recovery from full-page, camera-captured handwriting, with 537 page images (about 18.7k words, 163k strokes) and pseudo-label references filtered by OCR agreement and human raters. Models get a page image and a target word, then output ordered stroke points. Seven VLMs were scored with DTW, Chamfer distance, readability and style similarity. Closed-source models, notably Gemini-3.1-Pro, did best, and the ranking held on IAM-OnDB.

evals, multimodal, research· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Deogen: Automatic Benchmark Generation for Reasoning over Policies

DeoGen is an automatic, parametrizable benchmark generator for deontic reasoning, where a model applies policy rules to facts to reach a verdict. It autoformalizes rules into first-order logic, builds decision trees, and turns each tree path into a benchmark entry. Frontier models give the correct verdict 89-92% of the time, but only 22-40% on entries lacking enough information, even when told to abstain.

evals, research· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

AskStream: When and What to Ask Proactively under Uncertainty in Streaming Video Understanding

AskStream is a multi-turn benchmark for streaming video understanding that adds a third proactive action, ask, alongside wait and answer, for cases where video or user information is incomplete. It has 438 videos, 2,714 manually annotated frame-level time windows (wait, ask, answer), four task categories and eight subtasks. The strongest model reaches only 52.18% on the proposed Ask-F1 metric, in both multiple-choice and open-ended settings.

evals, multimodal, research· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Quantize by Doing: Task-Guided Mixed-Precision Quantization of Vision-Language-Action Models

QbD-VLA is a mixed-precision post-training quantization method for vision-language-action models. It uses task success from short closed-loop evaluations to assign weight and activation bit widths, accounting for interactions between components. On four VLA models across all four LIBERO suites it holds task performance at about 4-bit weights and under 6-bit activations. The gain on SIMPLER is small. On a Franka Research 3 arm, X-VLA shows no detected loss, but π0.5 fails.

research, inference, multimodal· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

When LLM Architectures Meet Smart Contract Vulnerability Detection: a Replication Study

This paper replicates Chen et al.'s study of LLM-based smart contract vulnerability detection, first on GPT-4 and then on Mixture-of-Experts, RNN-like and State Space Models. The exact replication confirms the high-recall, low-precision pattern on GPT-4. The differentiated replication finds the pattern does not transfer uniformly: GPT-4 and Gemma-3 behave alike despite a roughly 100x parameter gap, while two MoE models diverge.

research, evals· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Think Before You Act: Decision Gate for LLM Agents

JevAct is a runtime verification framework that intercepts selected LLM agent actions before execution and returns feedback for bounded revision. It compiles domain policies into executable state checks and structured questions scored by a prefill-only decision model, giving probability-based judgments without generation. Tested on eleven agents with tau^2-bench Airline and Telecom and BFCL, it gained consistently on Telecom and BFCL and was mixed on Airline.

research, agents, evals· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Repairs for Whom? Evaluator-Executor Mismatch in Selecting Repairs for Failed LLM Agents

Repairs for Whom? studies single-step repairs of failed LLM agent trajectories and asks whether common evaluators (greedy replay, stronger models, oracles, log-reading judges) pick repairs that help the deployed agent. Using independent rollouts in ALFWorld and WebShop, greedy replay loses 0.06-0.13 in expected gain, and Qwen2.5-14B rollouts misselect repairs for Qwen2.5-7B. The authors recommend verifying repairs with the executor that will run them.

research, agents, evals· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Data Scaling Laws in Stereo Matching: From In-domain Fitting to Cross-domain Generalization

This paper studies data scaling laws in stereo matching and releases XStereo, a realistic synthetic dataset of 200 thousand samples. In-domain training on five large datasets shows two regimes: a standard power law, then saturation, with a transition point roughly fixed for a given model. Cross-domain performance depends only weakly on source data scale, and distribution and coverage diversity reduce domain gaps more.

research, training, multimodal· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Session-Bound Behavioral Watermarking for Autonomous LLM Agents

AgentAnchor is a session-keyed behavioral watermark that lets an agent owner prove ownership of a trajectory log. Each step key comes from a secret key, a fresh public session identifier and semantic states of recent observations. Parity evidence is embedded through a KL-bounded exponential tilt of action choices. A trusted verifier checks disputed logs without rerunning the agent, using wrong-key calibration and abstention. The paper also formulates a learned cross-session forgery threat.

research, agents· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

CALM: Conflict-Aware Lifecycle Memory Governance for Long-Horizon Language Agents

CALM is a conflict-aware lifecycle memory governance framework for long-horizon language agents. It turns records into evidence-grounded memory cells, updates their validity and lifecycle states from support, conflict and dependency signals, and admits evidence at query time through training-free or learned gating based on temporal applicability. It beats baselines on LongMemEval-S and LoCoMo with Qwen and GPT backbones, and keeping superseded records helps earlier-state reasoning.

agents, research· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

From Leniency to Suspicion: How an Open Post-Training Recipe Changes Whom LLM Fact-Checkers Believe

This paper measures job-based doubt in LLM fact-checkers: judging a claim false more often when the poster has a lower-status job. It compares base and post-trained checkpoints of six model families on fact-checked claims with randomised poster profiles. Only the open Tülu-3 recipe (OLMo-2 32B, Tülu-3 70B) creates the effect, under one percentage point but reliable. In OLMo-2 32B the flip comes from one residual-stream direction and appears at supervised fine-tuning, confirmed on 400 new claims. None of seven commercial models doubts lower-status posters more.

research, ai-safety, training· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Benchmarking decision models against generative large language models on medical multiple-choice questions

This preprint benchmarks 13 decision models against 15 generative LLM configurations, including different reasoning settings, on 1,273 text-only four-option MedQA questions. Accuracy ran 67.71% to 89.08% for decision models and 78.32% to 96.86% for generative configurations. Decision models were cheaper and faster, but answers with probability of at least 0.90 still had error rates of 1.71% to 12.77%, and probability repeatability varied across three runs.

evals, research· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Does Chain-of-Thought Reliance Follow a Scaling Law?Larger Models Follow Their Steps More, Independent of Post-Training

This paper measures how much a model's final answer depends on its own written chain of thought, using a synthetic box-swap task with exact ground truth. It edits one step in the model's chain to a wrong but valid state and checks whether the answer follows. Across 25 open-weight models, reliance rises with size and saturates (Qwen3 from 16-18% at 0.6B to 97-100% at 14B). Reliance tracks capability (r = 0.99) and is unchanged by post-training.

research, ai-safety, evals· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

TCC-RAG: Temporal Causal Chain-Guided RAG for Temporal Knowledge Graph Question Answering

TCC-RAG is a retrieval-augmented generation method for temporal knowledge graph question answering (TKGQA). It turns the temporal knowledge graph into a multi-view causal cue graph, builds a query-specific temporal causal chain to guide retrieval, and applies a causal evidence filter to the results. The authors report gains of 10.72% and 8.40% on two TKGQA benchmarks and 62.97% on multi-step temporal reasoning.

research, llm-apps· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

FlatCake: A Flat KVCache Storage Architecture for LLM Serving

FlatCake is a KV cache storage architecture for LLM serving that treats host DRAM and NVMe SSD as peer media in one 64-bit address space, instead of a DRAM-over-SSD tier. Blocks are placed once, in session-contiguous, layer-major order, and streamed via GPUDirect Storage. On a Claude Code trace with DeepSeek-V4-Flash it holds a 94% hit rate (Mooncake: 64%) and cuts peak-load mean TTFT 1.6x.

inference, research· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Multi-Token Prediction Done Right

Soft Multi-Token Prediction (Soft-MTP) is a one-line change to multi-token prediction (MTP) pre-training. The authors show the MTP cross-entropy gradient splits exactly into a KL self-distillation term and a noisy residual, and a straight-through "KL gradient-recovery trick" keeps only the KL term. From 167M to 1.2B parameters it gives lower NTP perplexity, faster speculative decoding and higher zero-shot accuracy than standard MTP, with no added parameters or overhead.

research, training· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Complementing Text Based Moderation with Multimodal Grounding: Detecting Fraudulent Inducement Behaviors on Short Video Platforms

The paper studies detection and temporal localization of fraudulent inducements (prompts to follow, join chat groups, click links) in internet finance short videos. It compares OCR+LLM, OCR+MLLM and a zero-shot MLLM, and builds a dataset of 2,419 videos (66.49 hours, 3,194 events, four categories). OCR+MLLM reaches mIoU 26.52 versus 14.49 for keyframe OCR+LLM, with 67.0 percent macro recall, using a recall-weighted F2 localization reward.

research, multimodal, training· Added 30 Oct 2026, published 28 Oct 2026

huggingface.copaper

What Effort Level Should I Choose? On the Economics of Reasoning Effort in LLM Agents

This paper treats LLM reasoning effort as a credence good and measures the counterfactual effect of effort levels on proprietary and open-weight models across four benchmarks, including long-horizon agentic tasks. Higher effort often fails on tasks lower effort solved, and on easy tasks the top level costs up to 42x the lowest. Starting at the lowest effort and escalating only after a verified failure matches or beats fixed top effort, usually at lower cost for proprietary models.

research, evals, agents, ai-policy· Added 28 Oct 2026

arxiv.orgpaper

Post-Training Large Language Models with LLMON

LLMON (LLM Object Notation) is a framework that separates executable instructions from passive data and requires explicit identifier-based execution. The paper post-trains Granite-4.1 3B and 8B base models on a 4.6B-token corpus of general, RAG and tool-calling data with distractor variants, using SFT then RL. Attack success rates on indirect prompt-injection benchmarks fall from 53-100% to 0.02-5.6%, with general capabilities preserved. It also adds attention-based per-layer diagnostics.

training, research· Added 30 Oct 2026, published 28 Oct 2026

arxiv.orgpaper

Improving the Dependency on Fine-Tuning Data in Transfer Learning for Speech Emotion Recognition

This paper proposes an fMLLR-based compensation mechanism that cuts the fine-tuning data needed in transfer learning for speech emotion recognition under train-test acoustic mismatch. An autoencoder reduces trainable parameters, and an affine transform in the encoder latent space models rotations and shifts. In a human-robot interaction task, mean CCC rose 16.8% with 12.9% of the training set and 14.4% with 50%.

research, multimodal, training· Added 30 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Statistical Detection of Positional Bias in LLM-Based Recommender Systems: Methodologies and Empirical Evidence

The paper proposes POB-ICM (contingency matrix with Chi-square independence testing) and POB-UNBP (a uniformity null model under permutation-based evaluation) to detect positional bias in top-1, single-turn LLM recommenders. On a RAG product-recommendation testbed with 2000 queries and 20 permutations each (40,000 instances), both methods reject positional independence in all evaluated models. Three metrics quantify the effect, and the bias persists across candidate-set size and temperature. Code is public.

research, evals, llm-apps· Added 30 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Locating Sycophancy Across Training Stages in Open Language Models

Locating Sycophancy Across Training Stages in Open Language Models measures sycophancy in 13 open checkpoints from 5 lineages, from base model through supervised fine-tuning and preference optimization to release. It finds two behaviors with different origins: deference to a claim in the prompt exists in every base model, while capitulation under user objection is largely absent from base models and present in every post-trained checkpoint. Post-trained models often drop firmly held answers as readily as loosely held ones when the objection names an alternative.

ai-safety, research, evals· Added 30 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Heads of Security: Learned Attention Gates Improve Prompt Injection Robustness

Heads of Security (HoS) is a prompt injection defense that trains a small set of per-head attention gates to shift attention between trusted and untrusted tokens, separating instructions from data. Tested on Llama-3.3-70B-Instruct and GPT-OSS-20B, the black-box RL-Hammer attack reaches 10% ASR on HoS versus 97% on base GPT-OSS-20B and 85% on SecAlign. HoS keeps base utility in few-shot settings, where SecAlign and ASIDE degrade.

research, training· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Can LLMs Uncover Hidden Physical Laws through Free Exploration?

LongChang is an automated framework that builds simulated worlds with altered physical laws, such as heavier objects falling more slowly, to test whether LLMs can find the law by free exploration. Explorer LLMs manipulate objects and build tools without tasks or hints, while separate LLM agents generate scenes, review tool designs and grade reports. Across 12 LLMs, GPT-6 Astra reached a 16.7% discovery rate and the other 11 scored 0%.

evals, research, agents· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Scaling Laws for Tool-Use Agents via Controllable Environment Synthesis

SynthAPI is a pipeline that generates synthetic tool-use environments with explicit control over diversity of tools, tasks and underlying databases. Across 458,752 tasks and 849 trained models of four sizes, the authors fit a scaling law predicting downstream agent generalization. At larger data budgets, scaling tool and database diversity beats adding more tasks, and higher complexity helps more. Agents trained this way beat prior synthetic-data baselines on four benchmarks.

research, training, agents· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Metacognitive Decoding: Efficient Inference for Diffusion Language Agents

Metacognitive Decoding is a lightweight plug-in controller for diffusion language model backbones. It learns when to finalize tokens and how far generation should extend, and is trained on agent trajectories to balance correctness against generation cost. Across five benchmarks it cuts inference FLOPs by 20.6% on average versus the most efficient baseline, with competitive task accuracy. The abstract names no specific benchmarks or backbones.

research, agents, inference· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Know, then Draw: A New Learning Paradigm for Visual Generation

Know, then Draw is a learning paradigm for visual generation that uses autoregressive pretraining to acquire semantic knowledge and diffusion post-training for high-fidelity rendering. Controlled comparisons find autoregression the stronger pretraining objective, and semantic quality matters more than AR-only generation quality. This motivates a purely semantic tokenizer trained with language supervision only. A 3B text-to-image model trained in 6,600 H100 GPU-hours scores 86 on GenEval and 87 on DPG-Bench.

research, multimodal, training· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Perceiving and Predicting Dynamic Sports: Spatiotemporal Intelligence in Vision Models for Egocentric Tennis Video

This paper presents an annotated egocentric tennis dataset (egocentric video synced with five allocentric views, global positions, contact times, stroke types) and evaluates image encoders, video encoders and vision-language models on seven prediction tasks. Best models reach 0.14 m player and 0.54 m ball localization error and 0.14 s contact-time error. Depth input helps 3D ball localization, and longer histories do not help forecasting. A conditional diffusion model generates whole-body tennis motion in real time.

research, multimodal, evals· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

VideoLVR: Latent Visual Reasoning over Grounded Video Evidence

VideoLVR is a method that lets video multimodal LLMs build latent visual evidence from local regions at relevant moments and reason over it interleaved with text tokens. A visual token prediction objective, defined over a vocabulary derived from the input video, supervises the latents. It extends to reinforcement learning with sampled visual tokens as actions and final-answer rewards, and improves fine-grained video benchmarks without much context growth.

research, multimodal, training· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

EnvSched: Environment-Aware Scheduling for Rollout-Efficient Agent Reinforcement Learning

EnvSched is an online scheduler for agentic reinforcement learning that picks which environment-task pairs to roll out next. It builds a dual-relation graph over environments and tasks, propagates past rollout statistics to predict within-task reward variance, and adds environment quotas and a similarity penalty. On Qwen3-8B it scores 48.81 average on BFCL-v3, tau-Bench and ACEBench-Agent after 40 steps, above random selection at 120 steps.

training, agents, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

As the Story Unfolds: Watching a Film and Identifying Characters as a Human Does

This paper defines a character naming task for films: given a name spoken in dialogue and its video context, the model must pick which visible character bears that name. It provides a dataset with annotations for character naming and first appearances, a video-and-dialogue model, and a comparison with proprietary multimodal LLMs. It also proposes a progressive framework that builds a character bank as the film plays, without cast lists or actor photographs.

research, multimodal· Added 28 Oct 2026, published 27 Oct 2026

decrypt.coarticle

OpenAI Proves ETH Is False, and Crypto Twitter Has Questions

Decrypt reports that OpenAI published a 67-page paper (dated 17 October 2026, revised 25 October) claiming its AI found a deterministic subexponential-time algorithm for counting 3-SAT solutions, which would refute the Exponential Time Hypothesis. The proof was formalized in Lean. Impagliazzo, Paturi and Ryan Williams are reviewing it independently and say this is not certification. Anthropic posted a Claude-credited 3-SAT paper on 26 October.

research, models· Published and added 27 Oct 2026

arxiv.orgpaper

Interpretable Meta-Policies for Neural Decision Making

NEXUS is a hierarchical reinforcement learning framework that pairs neural skills with symbolic meta-policies, so transparent reasoning sits at the level of high-level skills rather than atomic actions. It uses object-centric observations and LLM interaction to design reward functions and meta-policies. Agents learn disentangled skills, align better with goals in the game manual, resist reward hacking more, and cope better with game simplifications than deep RL. The paper also compares neural, symbolic and neuro-symbolic meta-policy architectures.

research, training, ai-safety· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

CRISP: Co-designing Pruning and Scheduling for Efficient MoE Inference on Edge Servers

CRISP is a combination-aware framework that co-designs pruning and runtime scheduling for Mixture-of-Experts inference on memory-constrained edge platforms. It reorders neurons by importance so one stored expert supports several execution widths, gives hot, warm and cold expert combinations different widths, and overlaps expert loading with computation. On Mixtral-8x7B and DeepSeek-MoE-16B it raises throughput up to 2.1x at the same memory budget, with accuracy loss limited to 3.6%.

inference, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

FeatureArc: Benchmarking Coding Agents on Evolving Features

FeatureArc is a benchmark for coding agents on evolving features, built by a pipeline that mines repository histories into multi-stage tasks across four programming languages. It has 73 manually reviewed tasks of 2 to 10 stages, validated with fail-to-pass and pass-to-pass tests. Across 10 agents, pooled success drops from 62.7% at the first stage to 28.5% for the whole feature. Inherited defects appear in 72.9% of failed later stages.

evals, coding-agents, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

When Visual Evidence Is Insufficient: How Instructions, Post-Training, and Internal Activations Shape Refusal in Multimodal Language Models

This paper studies how multimodal LLMs refuse questions that an image cannot answer, using the MoHoBench benchmark on three MLLMs. It varies the meaning and placement of refusal instructions, compares supervised and preference-based post-training, and intervenes on a refusal-related activation direction. Instructions change refusal at the sample level in ways aggregate rates hide. Post-training raises refusal most but can mix refusal with attempted answers. Activation intervention gives smaller, transferable changes.

ai-safety, multimodal, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Bayes-Adaptive MDPs for Rollout Allocation to Enhance GRPO Training

AdaRoll-GRPO is a method that allocates GRPO rollout budgets per question by difficulty instead of uniformly. It frames allocation as a sequential decision problem and learns an allocation policy with a Bayes-adaptive MDP, computed offline so training adds no online optimization cost. Experiments on several reasoning datasets and LLM architectures show gains over standard GRPO at the same total rollout budget.

training, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

When Is Additional Retrieval Worthwhile? A Cost-Aware Study of Evidence Gain and Answer Reliability in RAG

This paper studies a cost-aware, two-action RAG routing policy that chooses between dense retrieval plus reranking and a hybrid lexical-dense retriever, using a regularized linear model to predict per-question evidence gain. On 60 held-out QASPER articles (188 questions), it triggers hybrid retrieval on 24.5% of questions and matches always-hybrid Hit@10 (78.70%). Answers from 4-bit Qwen3-4B show no clear correctness or faithfulness gain, and the assessments are model-assisted.

research, llm-apps, evals· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

LiveSearch: Questions as Executable Verifier Code for a Live Benchmark of Professional-Source Financial Search Agents

LiveSearch is a live benchmark for financial search agents that stores each question as executable verifier code: a typed program of calls to a professional financial data service plus a declarative extractor. Gold answers are recomputed daily with no human or LLM in the loop, so memorized answers go stale. Six tool-calling models score 70.9% to 76.7% under web search, and 70% of questions change gold value over three days.

evals, agents, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

OpenGray: a benchmark for evaluating the decisions of radiotherapy planning agents

OpenGray is a benchmark that adds tasks with known correct decisions to the OpenKBP-Opt head-and-neck dataset, including two overlap contradictions with provably incompatible dose goals and a feasible control. Four planning systems, from a rule-based script to a full LLM agent, were tested on 30 patients with Terra and Ministral 3 8B. Decision accuracy reached 95.4% and 92.1%, yet only 50% and 25% of overlap contradictions were escalated. SAGE logs of 39 radiosurgery runs showed an agent that lacked an action for stopping.

evals, agents, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Tessera: Structure-Completed Evidence Units for Web Retrieval-Augmented Generation

Tessera is a training-free evidence delivery layer for web retrieval-augmented generation. It rebuilds retrieved pages' HTML into fine-grained evidence units, each completed with page title, section path and table headers, then packs the relevant units into the reader context at the same token budget. Tracing evidence over full English Wikipedia shows chunking loses table and infobox content. Tessera adds 5 to 6 accuracy points on three QA benchmarks and carries over to agentic search.

llm-apps, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Exploring the Limits of the Delta Learning Hypothesis: Single-Pair Generator Biases in Direct Preference Optimization

This paper tests the delta-learning hypothesis for Direct Preference Optimization (DPO), where the larger model's response in a single model pair is treated as preferred. It shows, in theory and experiments, that fixed pairs add generator-specific biases that predict the preference label regardless of quality. DPO on single-pair data underperforms multi-model data, consistent with reward hacking, and adding more generators partly fixes it without an external judge.

training, research· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

p-DCIM: A 28nm Probabilistic SRAM DCIM based Ising Machine for Near-Memory Context Selection in LLM-RAG Pipeline

p-DCIM is a 28 nm SRAM digital compute-in-memory macro (0.16 mm2, 0.41 mW) that works as an Ising machine for RAG evidence selection. Selection is cast as a QUBO problem, and the same MAC datapath computes embedding similarities and annealing steps. Replacing LLM reranking with it gives a selection-energy estimate over six orders of magnitude lower and cuts total energy per query by 67%, with comparable token-level F1 on the evaluated examples.

research, inference, llm-apps· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

DCKV: Learned Transform Coding of the Prompt KV Cache in Hybrid Language Models

DCKV is a differentiable codec for the prompt KV cache of hybrid language models, which keep a KV cache only in their few full-attention layers. Transform, quantizer and entropy models (conditioned on prompt tokens and already decoded neighbouring positions) are trained end to end against a frozen model. On Qwen3.5-2B it stores a token in over 55 times fewer bytes than bf16 with at most 1.8 points lost on three long-context benchmarks, and over 100 times fewer with at most 3.6 points lost.

research, inference· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

AxisQ: Depth-Adaptive Channel--Frequency Quantization for Camouflaged Object Detection

AxisQ is a post-training quantization framework for camouflaged object detection (COD) that allocates activation bits by feature depth. It uses the channel domain for shallow features and the local-frequency domain for deep features, under a fixed per-layer average bit budget. It adds a boundary-aware Hessian approximation for reconstruction. On CAMO at W2A2, weighted F-measure improves 12.7% for CFRN and 33.3% for CamoFormer-P over the strongest baselines.

research, inference, multimodal· Added 28 Oct 2026, published 27 Oct 2026

eprint.iacr.orgpaper

OblivDGL: Oblivious Deep Graph Learning Made Practical

OblivDGL is a framework for secure GNN training and inference in machine-learning-as-a-service settings, modelled on Deep Graph Library. It uses an oblivious message passing paradigm and a new oblivious graph representation built on function secret sharing, revealing only the total node and edge counts. On four benchmark graphs it cuts training communication up to 1024x with up to 14.3x speedup, and inference communication up to 754x with 4.19x speedup.

research, training, inference· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper

Folding Attention Biases into Keys Makes Compacted KV Caches Run on Standard Kernels

The paper folds the per-key logit bias of Attention Matching (AM) KV cache compaction into the keys along one closed-form direction per head, fitted on context queries. The result is a plain cache that FlashAttention and vLLM paged caches can use, with output error bounded by how far each query's projection on that direction is from 1. Unmodified vLLM serves folded caches through a KV connector. On QuALITY and LongHealth the fold is within noise of AM in 13 of 14 settings. Under AM's published per-head budgets it is within noise in 9 of 10.

research, inference· Added 28 Oct 2026, published 27 Oct 2026

huggingface.copaper

Trial by Trace: Benchmarking AI Agent Harnesses for Reliability and Failure Attribution

TraceTrial-Bench is a benchmark that scores agent task performance and agent harness reliability separately. It has 95 mechanism-oriented tasks across seven components: context, execution, governance, lifecycle, observability, tools and verification. Across 4 harnesses, 9 models and 3,420 runs, macro-averaged harness scores range from 59.8% to 72.0%, and 31.1% of runs succeed despite failing at least one harness criterion.

evals, agents, research· Added 27 Oct 2026

arxiv.orgpaper

AK_QMUL at ArGuard Shared Tasks: A Single Fine-Tuned JAIS for Arabic Harmful-Prompt Detection

AK_QMUL's system for the ArGuard-2026 shared tasks fine-tunes JAIS-6.7B with 4-bit QLoRA and a linear head on the last-token representation. For binary safe/unsafe detection (B1) it uses a four-way head over dialect register and AraSafe MSA augmentation, reaching macro-F1 0.984. For seven-way harm domain (B2) it averages five seeds, reaching 0.790. Both ranked first. Bully is the weakest class.

research, ai-safety, training· Added 28 Oct 2026, published 26 Oct 2026

arxiv.orgpaper

Your Wish Is My Commit: Over-Compliant Refactoring in AI Coding Agents

Your Wish Is My Commit is a controlled study of over-compliant refactoring in Codex and Claude Code, run across five models, 50 algorithmic tasks and 10 object-oriented tasks. Over 3,000 turns of repeated "refactor further" prompts, no model stopped on its own: each rewrote up to 31x the original codebase before first stopping and resumed edits after declaring the task complete. The paper defines six behavioral signals for detecting over-compliance.

coding-agents, research, evals· Added 28 Oct 2026, published 26 Oct 2026

arxiv.orgpaper

Reva: Real-Time Vulnerability Detection through Contextual Probing of Large Language Models

REVA (REal-time Vulnerability Analysis) is a vulnerability detector that puts a module threat model, lightweight static analysis findings and call-graph information into one LLM forward pass. A trained probe reads the model's internal representations of the new function. On PrimeVul's held-out test set, REVA on GLM-5.3-Flash reaches 60.0% accuracy in about one second per function, matching the agentic VulTriage (59.2%) at over 250 times the speed and under a tenth of the cost. The probe is 14% to 17% more accurate than the LLM's own verdict.

research, coding-agents· Added 28 Oct 2026, published 26 Oct 2026

en.cryptonomist.charticle

Test-time training in robotics shows no clear gains in demo response

This article reports a study submitted October 23, 2026 by Qiangong Zhou, Jiafeng Liang, Yimin Dong, Yang He and Hongxiang Li. They added Test-Time Training (fast-weight memory) to the vision-language-action models InternVLA-A1 and Pi0.5, trained on paired human demonstrations and robot trajectories. Shorter videos, full-trajectory gradient connectivity and contrastive regularization gave no clear gain in demonstration-guided target selection at the tested scale.

research, multimodal, training· Published and added 26 Oct 2026

en.cryptonomist.charticle

A new AI method for saliency-aware image-text alignment improves image search

RoSA is a finetuning framework from Qiuyu Kong and three co-authors for long-text image-text retrieval. It targets "saliency collapse", where models over-rely on text tied to prominent visual features and miss contextual information. RoSA splits pairs into salient and contextual views using object-level saliency estimation and LLM text decomposition, then trains on both with plug-and-play auxiliary alignment objectives. Code and checkpoints are public.

multimodal, research, training· Published and added 26 Oct 2026

openai.comarticle

Working with academic communities to build scientific understanding

OpenAI describes three results its models produced in October 2026: a proof of the local smoothing conjecture in all dimensions, a disproof of the Exponential Time Hypothesis, and an existence proof for four-dimensional Yang–Mills and QCD. Each is shepherded by external expert groups under four principles (transparency, no academic credit, recognition, independence). Lean proofs and AI writeups for two results are on GitHub. All claims are OpenAI's.

research, ai-policy· Published and added 26 Oct 2026

arxiv.orgpaper

SymLift: Neural Semantic Lifting for Compile-Free Vulnerability Detection in Context-Incomplete Code Fragments

SymLift is a compile-free vulnerability detection framework for code fragments that lack project context. A fine-tuned LLM reconstructs control-flow, data-flow, memory and resource relations as a constrained LLVM-like IR, and a deterministic parser turns it into facts for Datalog rules that give findings with derivation traces. SymLift-Qwen-7B reaches 94.41% Target-F1 on FormAI and 91.66% micro-averaged Target-F1 on MegaVul.

research, training· Added 28 Oct 2026, published 25 Oct 2026

arxiv.orgpaper

Explaining Navigation Decisions: Stage-Aware Visual Attribution for Vision-Language Navigation

A stage-aware visual attribution framework for Vision-Language Navigation (VLN) that explains a robot's action choices. It identifies "instruction-stage insensitivity", where existing explainers give similar outputs across navigation stages. The method builds action-conditioned temporal instruction-visual attributions, aggregates relevant historical evidence by stage, and applies an instruction-guided spatial filter. It improves attribution quality across several VLN policies and benchmarks, helps diagnose failure cases, and was tested on a Unitree G1 robot.

research, multimodal· Added 27 Oct 2026, published 25 Oct 2026

arxiv.orgpaper

The Patient Is Always Right? Evaluating Sycophancy in Medical Diagnosis with Graded Patient Beliefs

This paper evaluates sycophancy in medical diagnosis across nine AI models. Each model is shown clinical evidence plus a patient's belief in a specific but incorrect condition, with the belief's closeness to the evidence varied. Models adopted the belief 23.2% of the time on average, even for fictional conditions, and 43.0% under a Rare Disease Claimer persona. Medical fine-tuning and larger scale reduced susceptibility.

ai-safety, evals, research· Added 27 Oct 2026, published 25 Oct 2026

arxiv.orgpaper

Reversible Pseudonymisation of Documents Sent to External LLMs: an Open Bilingual Benchmark and an Engine

This paper presents an open benchmark of five existing pseudonymisation tools and a regex baseline on Russian and UK English documents, plus an open engine built on the two strongest tools. On a held-out set of 1,000 synthetic documents per language, the best existing tools leave a whole value visible in 58 to 61% of documents. The engine does so in 0.2 to 0.3%. Russian name restoration reaches only 94.6%, below the 99% threshold. Real documents were not tested.

evals, llm-apps, research· Added 27 Oct 2026, published 25 Oct 2026

arxiv.orgpaper

Detecting and Understanding Ambiguity in Visualizations with Role-Playing Vision Language Model Interpretations

This paper presents a pipeline that uses persona-conditioned vision language models to surface interpretive ambiguity in visualizations. Topic-relevant personas elicit candidate takeaways, and a Target-Pattern-Scope framework lets LLM judges label statement pairs as Compatible, Disjoint or Opposing, which are summarized in an ambiguity profile. The authors benchmark VLMs on chart comprehension per chart type, check labeling against human annotations, and run case studies on public visualizations and a redesign.

research, multimodal, evals· Added 27 Oct 2026, published 25 Oct 2026

arxiv.orgpaper

InteractiveWorld: A Verifiable Human-Centric Environment for Building Better Interactive MLLM Assistants

InteractiveWorld is an environment for closed-loop interaction between an AI assistant and a simulated human agent who sees the scene, follows instructions and acts independently. InteractiveBench, built on it, has 120 multi-turn tasks in six categories scored by verifiable, state-based rewards. Across 12 frontier MLLMs, the best model, Claude Fable 5.1, reaches at most 20% on long-horizon Guidance and 25% on Planning tasks.

evals, multimodal, research· Added 27 Oct 2026, published 25 Oct 2026

arxiv.orgpaper

Continual Learning Exam: Benchmarking Long-Horizon RSI and Experiential Learning in LLM Agents

Continual Learning Exam (CLE) is a benchmark of 120 human-authored tasks, each a five-level ladder (600 levels) across 89 professional roles and 12 occupational domains, built by 50+ domain experts. Agents must pass a level to unlock the next, and an evaluation agent scores submissions against human rubrics. Across 17 harness-model configurations, feedback plus one revision adds a median 5.3 points; the best system clears 47.7% of levels and finishes all five levels in 23.3% of tasks.

evals, agents, research· Added 27 Oct 2026, published 25 Oct 2026

arxiv.orgpaper

HerHealthGPT: An Evidence-Grounded Conversational AI System for Women's Reproductive Health Copy title

HerHealthGPT is a conversational AI system for women's reproductive health that combines retrieval-augmented generation, context-aware clarification and source validation in a web interface. Evidence retrieval raises concern-category accuracy from 54.1% to 59.3% over fine-tuning alone on 540 messages from 90 cases. The clarification router reaches 85.6% recall and 86.2% specificity. In a 30-adult study, 83.3% preferred the retrieval-enabled version. The authors say clinical validation is still needed.

research, llm-apps· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

DRILS: Dependency-based Requirement Change Impact Analysis Using LLMs

DRILS is an LLM-based agentic framework for requirements change impact analysis. It uses multiple agents to generate candidate impact sets, with an LLM-as-a-judge promoting diversity in selection. It adds RE-RAG, a tuned retrieval technique, and a requirement dependency graph as a knowledge source. On three curated datasets it reaches 87.1% recall, 76.3% precision and 81% F1 at lower cost than baselines. The authors release the datasets.

research, agents, llm-apps· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Intervention-Certified Amplification: A Validity Gate for $σ$-Weighted Post-Training

Intervention-Certified Amplification (ICA) is a validity gate for sigma-weighted hybrid SFT/RL post-training. The paper shows that DIAA, a per-sample sigma-weighted hybrid, emits unexecuted tool-integrated reasoning on 29% of MATH-500 prompts (three times vanilla GRPO) even with Python fences removed from its anchor. ICA gates sigma with paired-rollout evidence, cutting TIR to 13.8% (frozen) and 11.2% (online) and raising accuracy from 47.5% to 48.7% and 49.3%. It transfers to GSM8K.

training, research· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Build, Don't Play: Self-Improving LLM Research Agents Develop a Hearthstone Battlegrounds AI That Rivals Top Players

AutoTavern is a self-improving system of specialised LLM research agents with a typed shared memory and verification gates. In 20 days it built TavernGym, a full-content Hearthstone Battlegrounds simulator, and 23 versions of TavernAgent, a PPO-trained player. TavernAgent averaged placement 1.60 with 80% first places over its last 20 live games, beating the top ten players (2.34, 59%). On TavernBench (960 tasks), it answered 91% of play tasks against at most 63% for frontier LLMs.

agents, research, training, evals· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Sliding Networks: A Family of Residual Structures Preserving Feature Learning and Feature Diversity

Sliding Networks is a family of residual network structures with a joint width and depth scaling recipe that keeps both locally maximal feature learning and feature diversity in the residual stream. The key quantity is the ratio N_flat/(N_sharp L) tending to a constant alpha, and Half Copy correlated initialization can relax the shape constraints. Residual-MLP experiments support an SDE limit, but language-model experiments show performance differences under regrouping at fixed parameter count.

research· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Learning What to Share: Expert Specialization for Time-Series Forecasting

Persistent Closed-Form Routing (PCR) is a plug-and-play method for multivariate time-series forecasting. It alternates exact conditional estimation of channel-specific mixture weights with gradient training of low-rank residual experts around a frozen pretrained forecaster. Across eleven datasets and three backbones, PCR-K4 cuts dataset-balanced MSE by 3.86-14.25% and beats competing specialization methods with fewer trainable parameters. The paper also introduces GalaxyTS, a synthetic dataset with exact shared and channel-specific references.

research· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Task Format as a Factor in Adaptation: A Controlled Comparison of LoRA and Full Fine-Tuning for Prompt-Safety Classification

This paper compares LoRA and full fine-tuning for prompt-safety classification using a matched 400M encoder-decoder pair (Ettin) across three task formats, with up to 15 seeds per cell and pre-specified analyses. Full fine-tuning leads LoRA by 2-3 pp at the pre-specified learning rates, and the encoder gap closes when both use the same learning-rate search. Training LoRA to write a reason after its verdict adds 4.0 pp, while a content-free sentence of equal length does not. Trained models beat a prompted 8B in domain but trail it by 8-23 pp on adversarial jailbreaks and over-flag benign prompts (27-66% against 2-3%).

training, research, evals· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

IntentDelta: A Human-Constructed Benchmark for Evolving User Intent

IntentDelta is a human-constructed multi-turn benchmark of 638 trajectories and 3,766 turns across WebShop and TravelPlanner. Each turn pairs a user utterance with a cumulative intention state and human reference actions. Across eleven agents, average Intention Recovery is 33.1% in TravelPlanner and 44.7% in WebShop, and average Action Score is 31.6% and 41.0%. Agents lose earlier requirements and confuse soft preferences with must-have constraints.

evals, agents, research· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Unlocking the Specialization Capability of PPG Foundation Models for Wearable Signals through Structured Pruning and Adaptive Fine-Tuning

SPAFT is a structured prune-then-finetune framework for PPG (photoplethysmography) foundation models used on wearable signals. It finds that pretrained PPG models have inherent sparsity and redundancy, then restricts fine-tuning to a compact parameter subspace chosen by parameter importance and structural redundancy. Across four PPG foundation models, five datasets and ten tasks, it wins 82.8% of regression and 87.5% of classification comparisons.

research, training· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

TriageRAG: Failure-Aware Self-Training for Adapting Retrieval-Augmented Language Models to Specialized Domains

TriageRAG is a failure-aware self-training method for adapting retrieval-augmented LLMs to specialized domains using only an unlabeled corpus. It probes the student on self-generated questions with and without retrieved context, sorts them into known, solved, context-failure and retrieval-failure, and routes each to a matching training signal. With Qwen3-4B on 11 datasets it beats budget-matched SimRAG by 1.2-6.4 points and recovers closed-book accuracy where retrieval hurts.

research, llm-apps, training· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

An actively trained neural operator for near-explicit atmospheric chemistry

FNO-MoE is a Fourier neural operator with a mixture of experts that predicts the full near-explicit chemical state (Master Chemical Mechanism, thousands of species) over 30 minutes in one call. Expert disagreement selects the most informative box-model simulations, so active sampling beats random or evenly spread sampling at equal cost. Fine-tuning on longer trajectories keeps multi-hour runs stable, and the emulator runs orders of magnitude faster than the numerical solver.

research· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Enhancing LLM-Based Vulnerability Detection via Representation-Aware Logical Ensemble

RALE (Representation-Aware Logical Ensemble) is a method for LLM-based vulnerability detection in C code. The authors train LLMs on raw source code, Abstract Syntax Trees and Control Flow Graphs, and find each representation works best for certain vulnerability classes. Combining the models with explicit logical rules and majority voting raises F1 from at most 73% for a single model to 78%, with gains in accuracy, precision and recall.

research, training· Added 28 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

ZipFusion: Fusing Lossless KV Decompression with Attention for Efficient LLM Decoding

ZipFusion is a lossless KV cache compression scheme whose attention kernel reads the compressed cache directly and rebuilds the original values in registers with simple bit operations, so no separate decompression pass is needed. It also reassigns the memory freed by compression across layers to hold more KV cache. Built on FlashInfer and integrated with SGLang, it gives about 1.25x decode attention speedup on an H100 and up to 26.3% higher decode throughput at 8K context than uncompressed SGLang. It reaches up to 2.76x the throughput of the best lossless KV compression baseline.

inference, research· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

VulDec: Vulnerability Detection in Binaries via Neural Decompilation

VulDec is a dataset and framework for detecting vulnerabilities in binaries by first decompiling them with fine-tuned LLMs. The dataset aligns real-world Assembly-C function pairs across several architectures with CWE labels. Tuned neural decompilers reach about 56% Edit Similarity and 43% BLEU-4, and tuned detectors reach a macro F1 of 0.61 on a realistic test set with a 3% vulnerability rate.

research, training· Added 28 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

RAQ-MLA: Role-Aware Quantization for Multi-Head Latent Attention Caches

RAQ-MLA (Role-Aware Quantization for Multi-Head Latent Attention Caches) is a training-free method for quantizing the MLA KV cache. It fits a separate invertible transform to the compressed latent (tuned for attention-weighted value error) and to the positional key (tuned for score perturbation). It reaches about 2.5 bits per element, a 6.5x gain in theoretical KV-cache capacity, with near-lossless reasoning and coding results on four MLA models.

research, inference· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Eat Your Precision: Precision Objective Mitigates Reward Hacking in Rubric-based RL

FORK is a reward for rubric-based reinforcement learning that jointly optimizes recall and a new precision objective. Precision requires each unit of a response to be necessary (rubric-driven necessity credits) and factually correct (a grader). Rubric-coverage rewards otherwise produce verbose, less accurate responses. FORK keeps similar coverage with shorter outputs, raising win rate against the base policy from 10.6% to 68.0% on RubricHub-Medical and from 22.1% to 74.1% on HealthBench.

training, research· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

DualPace: Decoupled Scheduling and Cooperative GPU--CPU Decoding for LLM Serving

DualPace is an LLM serving runtime that treats the host as an independently scheduled decode engine. A GPU engine runs prefills and GPU-resident decodes, while an AMX-enabled CPU engine batches and advances CPU-resident decodes, optionally calling GPU Dense Assist for dense stages. In decode-oriented scenarios it raises throughput 6.9 to 43.2% over NEO and 11.5 to 90.6% over vLLM at similar latency.

inference, research· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper

Small Edits, Shifting Decisions: Benchmarking LLM Accuracy and Reliability on French Input Variants

This paper builds a controlled French surface-variant benchmark with seven task types and 4,200 original-low-high triplets (12,600 inputs), covering missing apostrophes and diacritics and keyboard typos. Six models from two LLM families lose accuracy and show worse Brier score and NLL as perturbation grows. Keyboard typos hurt most. Calibration temperatures fitted on original inputs transfer to variants, and combining them with input restoration improves reliability further.

evals, research· Added 27 Oct 2026, published 24 Oct 2026

huggingface.copaper

IQuest-Q1: Advancing Foundation Capabilities for Agentic CLI Systems

IQuest-Q1 is an open-source sparse Mixture-of-Experts model with 320B total and 15B activated parameters, built for agentic coding in CLI systems. Pre-training combines Stable SplitNorm, Muon and muP with a two-stage curriculum that raises the coding-data share in stage two. Mid-training extends context, and post-training mixes SFT, domain-specialized RL and MOPD across scaffolds, followed by model merging.

models, training, coding-agents, research· Added 24 Oct 2026

arxiv.orgpaper

Hey, guess what: CUES lead to inferences about typicality

The paper builds two corpora, EventTyp and the human-validated EventTypGold, to study how people infer typicality and atypicality of events from what is said. Large-scale human and LLM experiments show that three CUES predict these inferences in humans: Commonsense Knowledge, Utterance Emphasis and Snark. In most settings LLMs lack the commonsense knowledge to judge an event's general typicality, so they struggle to follow the CUES.

research, evals· Added 27 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Deletion, Substitution, Repair and Transplant: Component Attribution in BARBICAN, a Layered Prompt-Injection Gateway for MCP

The paper analyses BARBICAN, a fail-closed prompt-injection gateway for MCP tool mediation, using deletion, substitution, repair and transplant experiments. Its arbitration classifier ran on tokenizer-mangled text for a whole evaluation. Fixing that raised AUC from 0.392 to 0.855 while system recall fell from 80.67% to 80.34%. The band it receives is 96.82% malicious. The authors argue ablation figures need the band's class balance and substitution baselines.

evals, research, mcp· Added 27 Oct 2026, published 23 Oct 2026

huggingface.coarticle

LitPivot: Developing research ideas through dynamic contextualization within the literature landscape

LitPivot is a system from Ai2 and the University of Pennsylvania that helps researchers revise early research ideas when reading prior work changes their direction. It splits a proposal into problem, contribution and evaluation facets, groups papers from Asta Paper Finder per facet, and offers critiques with supporting excerpts and suggested rewrites. In a 17-researcher study, domain experts rated LitPivot ideas as better grounded in the literature than ideas made with a text editor and an AI chat tool.

research, llm-apps· Published and added 23 Oct 2026

arxiv.orgpaper

AI Alignment for Digital Democracy: Fairer Voters Representation against LLM Biases

This paper studies demographic bias in LLM-emulated electorates, using three U.S. presidential elections and comparing one closed-source and one open-source LLM. Removing gender from the voter profile cuts the largest group error by 22-42%, and removing education cuts it by 64%. Removing self-reported race raises mean group error by over 40%. Repeated queries and fairness instructions do not reduce errors and can double them.

research, ai-safety, evals· Added 27 Oct 2026, published 23 Oct 2026

latent.spacearticle

AI Is Getting Good at Math. Here’s Why.

The essay argues that LLMs solve open math problems because they search depth first, and automated verifiers make failed attempts cheap. It uses WT Gowers' hint-sequence complexity (2^k + t), adds a branching factor b, and walks through OpenAI's unit-distance disproof and Cycle Double Cover prompt as a drafter/solver/verifier harness. It predicts AI does best where k and b are small.

research, agents· Published and added 23 Oct 2026

arxiv.orgpaper

What Transfers Across Protein Pockets? Experience Reuse in Molecular Optimization for Agentic Structure-Based Drug Design

This paper studies experience reuse for LLM agents that optimize ligands in protein pockets by editing and docking. Using the full log of one agent run over CrossDocked2020 test-set pockets, it tests whether keys in an experience base recur on unseen pockets and separate improving from worsening edits. The separation is largely explained by molecule size change, action and round, not the target. The authors conclude a base may better prioritize an agent's own proposals for docking than prescribe what to build.

research, agents· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

OWLS: Open-world Language-conditioned Speed Fields for Zero-Shot Rover Navigation

OWLS (Open-world Language-conditioned Speed Fields) plans directly over a vision-language model's dense feature map. A text prompt describing traversable terrain scores every pixel of one camera image. The scores are projected onto the ground, adapted with two morphological operations, and used as the speed field of a minimum-time planner. It needs no training, terrain labels, map or extra sensors. It was tested on a physical rover, on simulated Mars and Moon rovers over 200 procedural terrains, and on real planetary imagery.

research, multimodal· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

PHYRD: A Living Benchmark for Reproducing and Extending Theoretical Physics Research

PHYRD is a living benchmark for theoretical-physics reasoning with 63 Reproduce problems (derive published results) and 43 Discover problems (consequences of modified theoretical conditions). Tasks are built by FormulaForest, with reference solutions and subquestion-level rubrics. Across five models, whole-problem accuracy is 49.21% to 68.25% on Reproduce and 18.60% to 74.42% on closed-book Discover. FormulaForest data also lifts a Qwen-35B-A3B model on FrontierScience from 61% to 72%.

evals, research· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Standing on Open Shoulders: Regional Fine-Tuning of Open-Weight Camera-Trap Models for UK Wildlife with Modest Data and Compute

DeepFaune-UK is a fine-tune of the open DeepFaune camera-trap classifier on 28 UK classes, paired with MegaDetector v6. On 20,212 MammalWeb images it reached 88.8% strict accuracy against 56.5% (DeepFaune v1.5.0) and 59.0% (SpeciesNet). Re-applying the recipe to the DINOv3-based classifier gave 90.6%. Fine-tuning on about 150 crops per class matched full-data out-of-distribution accuracy, and a DINOv2 run took about 22 minutes on under 4 GB of GPU memory.

research, training, multimodal· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

UAV-SwarmAgent: Scalable Multi-UAV Reconnaissance and Task Allocation via Code Generation and Vision-Language Reasoning

UAV-SwarmAgent is a hierarchical multi-agent system where an operator gives one natural-language sentence and a ground-station agent writes and runs code that produces per-UAV routes. A vision-language model grounds targets in an aerial map, and worker agents verify incidents with a compact onboard model. In simulation it located the incident in 244 of 255 missions (95.7%), and flight time followed an Amdahl-type law with speedup bounded at 1.52x for one to five UAVs.

research, agents, multimodal· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Tool-Calling LLM Agents Under Realistic MCP Faults: Scale Does Not Purchase Fault Tolerance

MCP-Injector is a protocol-aware fault-injection middleware for Model Context Protocol tool calls, with permanent, transient and intermittent faults on seedable schedules. Its fault model comes from 57,240 issue and PR threads across 1,774 open-source MCP repositories. In a 1,350-cell experiment, faults cut task-completion score by 60% (8.08 to 3.23 of 10), and the 120B gpt-oss model was no more resilient than the 20B. Agent stacks differed in how well they absorbed faults.

mcp, evals, agents, research· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

DRIVE: Fine-tuning via Data Contribution- and Diversity-aware Weighting with Prior Regularization

DRIVE is a per-sample weighting framework for single-stage fine-tuning when the source pre-training data is unavailable. It uses one differentiable objective: Shapley additivity splits each sample's contribution into anti-forgetting and adaptation parts, a covariance-based diversity term reshapes them, and a sample-characteristic prior stabilizes them. The authors prove better stability-plasticity guarantees than existing weighting schemes and report better performance, faster convergence and more robustness.

training, research· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Accordion: Elastic-Depth Inference via Fixed-Prefix Layer Looping

Accordion is a mixture-of-experts language model (Accordion-6B-A0.5B) that supports five inference budgets through fixed-prefix layer looping, with the minimum budget being the unlooped model. After 3.6T pretraining tokens, deeper inference raises the macro-average over 37 tasks by 3.01 points, and the middle budget captures 89% of that gain with half the extra layer executions. SFT and RL keep the budgets, merging three domain experts via parameter-group merging and budget-matched multi-teacher on-policy distillation.

research, training, inference· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Efficient Zero-Shot Video Anomaly Detection via Signal-Aware Adaptive Fusion (SAAF)

Signal-Aware Adaptive Fusion (SAAF) is a training-free zero-shot video anomaly detection framework that cuts the compute cost of vision-language models. It combines a dual-signal sampler (frame-to-frame delta and rolling-context drift in embedding space, OR-gated with EMA thresholds), a threshold-gated cascade from frame-level to video-level analysis, and signal-aware score fusion. It reports AUC-ROC of 92.07% on XD-Violence, 86.70% on MSAD and 80.54% on ShanghaiTech, at 2.2x to 2.6x the throughput of the prior best zero-shot method.

research, multimodal· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

When Agents Fail: Model or Harness--Who's to Blame?

When Agents Fail: Model or Harness--Who's to Blame? proposes five measures for evaluating agent harnesses together with their underlying models: Valid Completion Rate, Model-Harness Fit, Cross-Model Fit, Interaction Residuals and Token Load. It tests four harnesses (OpenCode, Codex CLI, Pi CLI, DeepAgents) with four models on six benchmarks, across 111 settings and 52,168 trajectories. Harness rankings change by model and benchmark. OpenCode ranks first in cross-model performance.

evals, agents, research· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Sharpness-aware Orthogonalized Momentum Algorithms

SOMA (Sharpness-aware Orthogonalized Momentum Algorithm) combines sharpness-aware minimization (SAM) with the Muon optimizer for LLM training. It avoids SAM's extra gradient computation by using stale, purely stochastic gradient noise as the adversarial perturbation. The authors prove sample complexity does not depend on perturbation direction, study robustness to heavy-tailed gradient noise, and test SOMA against post-training quantization degradation and catastrophic forgetting during fine-tuning.

research, training· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

SCI-AGENT-EXAM: Benchmarking Agents on Scientific Research Tasks

SCI-AGENT-EXAM is a workflow-oriented benchmark of 120 expert-curated tasks across 11 scientific fields and 7 task types from the Scientific Workflow Task Taxonomy (SWTT). Tasks need both executable and textual artifacts, scored by programmatic checks plus an expert-calibrated LLM judge. Ten frontier agents were tested: GPT-6 Astra scored highest at 80.9, but only 26.0% of its runs passed the 0.95 success threshold.

evals, agents, research· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Profile-Driven User Simulation via Reinforcement Learning with Hierarchical Rewards for Task-Oriented Dialogues

ProfileSim is a profile-driven user simulator for financial marketing outbound calls. It models user profiles (demographics, Big Five traits) and business profiles (intents reverse-engineered from real call logs). The base model is fine-tuned on real voice dialogues, then refined with GRPO using dialogue-level and utterance-level rewards. Against other user generators it scores highest on colloquial naturalness and profile adherence, and lowest agent score on long-tail intents, with shorter utterances at comparable business compliance.

research, training· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

When to Stop Penalizing Actions? Martingale GRPO for VLA Reinforcement Learning

MART-GRPO is a variant of group-relative policy optimization for vision-language-action (VLA) reinforcement learning. It uses a LoRA-fine-tuned VLM to find the last visible error without recovery in a failed rollout video, keeps failure feedback up to that point, and drops it afterward. Martingale stopping conditions show this preserves the expected GRPO gradient and can reduce covariance. With OpenVLA-OFT on RoboTwin 2.0, it raises success rates over full-trajectory GRPO across multiple tasks.

research, training, multimodal· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

MAST-VLA: Skill-Conditioned Adaptation with Trajectory-Affordance Priors for Multi-Skill Robot Manipulation

MAST-VLA is a vision-language-action framework for multi-skill robot manipulation that conditions on both the skill to execute and where and how to contact objects. It adds skill-conditioned LoRA Mixture-of-Experts adapters to the LLM backbone and turns a predicted affordance heatmap into a patch-level spatial prior for the action head. It reaches an Avg. Len of 4.65 on CALVIN ABC→D and was tested on a real robot.

research, multimodal· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

BiLoCo: Binary Low-Rank Corrections for Post-Training Quantization

BiLoCo is a post-training quantization method for FP4 LLMs that adds low-rank corrections whose left and right factors are stored as binary ±1 sign vectors instead of BF16. This uses 16x less storage per rank-one component and lets the correction run on CUDA cores alongside the FP4 tensor-core GEMM. On 4B to 72B LLMs it matches or improves memory-matched SVDQuant. A fused B300 kernel is at most 11% slower than FP4-only GEMM at batch size 1 for ranks up to 512.

inference, research· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

MPQ: A Message-Passing View of Post-Training Quantization

MPQ (Message-Passing Quantization) is a post-training quantization method that treats row-wise PTQ as posterior inference over discrete codebooks. It keeps soft posteriors across weight blocks using vector approximate message passing, then refines a hard-decision initialization with exact block coordinate descent. Tested at 2-bit lattice and 3- and 4-bit scalar quantization on Llama, Qwen and Mistral models, it improves perplexity and zero-shot accuracy over GPTQ-style baselines.

research, inference· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Semantically Rich, Metrically Affect-Blind: A Geometric Blind Spot in Modern Text Embedders

Sentiment-Injected Sentence Embeddings (SISE) addresses a finding across eleven frozen text encoders (2021 to 2026): linear probes recover sentence polarity at up to 91.3% accuracy, but cosine similarity separates polarity by only 0.003-0.064. A small supervised-contrastive head (0.13M-0.69M parameters) trained on CPU raises retrieval MAP by up to 26 points, but the 128-d head loses 38-41 STS-B Spearman points. The paper also releases the Attribute Alignment Gap diagnostic.

research, llm-apps· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

Tampered ML: Evaluating LLM Judgment of Machine Learning Research

Tampered ML is a benchmark of 1,858 item-pairs, each pairing an experiment proposal reconstructed from a published ML paper with a minimally edited version that adds a targeted methodological flaw. Models must pick the sounder proposal in both presentation orders. A human study estimates 86% of items are valid (95% CI 81-91%). Opus 5 reaches 77.9% accuracy, and models are sensitive to order and favor flawed designs that look more specific, transparent, or conventional.

evals, ai-safety, research· Added 26 Oct 2026, published 23 Oct 2026

arxiv.orgpaper

How Much Do You Trust A Dog? Quantifying Hint-Following in Reasoning Language Models

This paper measures how hints such as "[Source] thinks the answer is (X)" shift LLM answers on MMLU questions, varying the source between a professor, a friend and a dog across Qwen, Gemma and SmolLM models. The shift fits an exponential tilt: a roughly constant mean change in the hinted answer's log-odds, with susceptibility largely shared across sources. The shift persists when the source is declared inaccurate or the prompt warns against trusting it. Synthetic experiments show how KL-regularized reward training could produce this pattern.

research, ai-safety, evals· Added 26 Oct 2026, published 23 Oct 2026

huggingface.copaper

Cache Engineering for Agents (AKV)

AKV introduces cache engineering, which lets LLM agents control their active KV cache with drop (exclude cache from future attention) and repos (adjust retained keys' position encodings) without editing the trajectory or re-prefilling. The AKV serving engine keeps a stateless API while managing stateful cache reuse, eviction and restoration. A tool-response rolling-drop policy raised benchmark scores by 1.7 points and cut cost 27.7% versus threshold-triggered summarization.

inference, agents, research· Added 23 Oct 2026

anthropic.comarticle

LLMs are making formal software verification feasible at a much larger scale

Anthropic and Theorem report that Claude wrote Rocq proofs of crash-freedom for 239 OSS-Fuzz entry points across 123 open-source libraries, working on compiled RISC-V binaries for inputs up to 64 MB. Proofs total about 180 million lines, and Claude found and patched bugs along the way. The post covers proof reuse (17% fewer tokens), Opus 5.5 using about 40% fewer tokens than Opus 5, and limitations.

research, coding-agents· Added 23 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Sensorimotor Alignment: A Platonic Proxy for Embodied Policy Success

Sensorimotor Alignment (S^2) is a training-free offline metric that measures how well an embodied model's sensory representation aligns with the geometry of physical motion. It extends the Platonic Representation Hypothesis to motion trajectories. S^2 correlates positively with downstream robot task success across embodied model families, helps pick promising VLM backbones before fine-tuning into VLAs, and explains behaviours seen in VLAs and world action models.

research, evals, multimodal· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Causal Chain of Thought: Mitigating Attention Leakage via Context Isolation

CausalCoT is an inference framework that splits a problem into a directed acyclic graph of atomic sub-questions and evaluates each node in an isolated context holding only its direct causal parents. This removes attention paths to non-parent content, so intermediate steps drive the answer rather than leaking from global context. It keeps benchmark accuracy competitive and raises causal sensitivity of reasoning steps by 12 points on average, up to 21, over the strongest baseline. The graph also allows test-time corrections that propagate downstream.

research, prompting· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

BeNice! Closing Looping's Inference-Memory Gap via KV-Sharing

BeNice! is a post-training method for looped language models that broadcasts the final recurrence's KV-cache into the first recurrence, so one cache can serve all recurrences. It follows a first stage that uses stochastic depth in pretraining to induce contractive residual dynamics. It matches standard finetuning's unshared validation perplexity and composes with multi-latent attention, Kimi delta attention and fp8, mxfp8 and nvfp4 cache quantization. In simulated low-memory long-context inference it raises cache hit rates by up to 31.4% and throughput by up to 23.6%, and cuts p90 time-to-first-token by up to 83.6%.

research, inference, training· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

F2G-Pose: Geometry-Aware Foundation Feature Lifting for Direct RGB-D Category-Level Object Pose Estimation

F2G-Pose is a direct, single-pass framework that predicts category-level object pose, metric size and camera-frame completed shape from one RGB-D image. It lifts frozen RADIO foundation features onto partial point clouds, tokenizes local geometry with DGCNN, and fuses it with a Mixture-of-Experts Transformer. It trains only on synthetic SOPE data, leads on SOPE and ROPE at 28 FPS, and is tested on HouseCat6D, HANDAL and real manipulation.

research, multimodal· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Progress Probes $P^2$: Lightweight Runtime Monitoring for Vision-Language-Action Models

Progress Probes (P^2) trains linear probes on a vision-language-action model's own activations, using demonstrations, to estimate task progress, then flags executions that deviate from expected progress. Tested on the open-source pi0.5 across simulated and real-world manipulation tasks, it matches or beats VLM-based monitors and failure-data-trained methods at failure detection, without an external foundation model.

research, multimodal· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

How Ground Truth Shapes Retrieval Scores: Evidence-Aware Evaluation of a Retrieval-Augmented Generation System for Breast Lesions

This paper describes an evidence-aware retrieval evaluation protocol for a clinical RAG system over breast lesion management literature. Relevance is defined by evidence spans mapped to all containing chunks. A hybrid retriever (dense, BM25, cross-encoder reranking) with 384-token chunks and 20% overlap reached 0.94 Recall@3 on 18 queries. Expanding ground truth from 25 to 37 spans did not change rankings. The authors note the small set needs independent validation.

llm-apps, evals, research· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Who Is Sycophantic to Whom? Confident LLMs, Human Deference, and the Collapse of Trust Calibration

This paper studies reverse sycophancy: whether users bend toward a confident, consistent, wrong LLM. In a within-subjects experiment, 32 participants learned science topics from five GPT-5-mini assistants with 0 to 100% of facts replaced by falsehoods. Scores fell from 7.4 to 2.9 out of 10, planted falsehoods were chosen on 60% of falsified items against 11% of correctly taught ones, and confidence stopped tracking correctness. It introduces a baseline-corrected Human Deference Index.

research, ai-safety, evals· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Rethinking Sequence Compression with Test-time Regression

TTR-Compressor is a sequence compression method that treats input tokens as a test-time regression problem. The regression is learned end-to-end by backpropagation, and its coefficients are solved in closed form in the forward pass to give the compressed representation. As a block-wise sparse attention compressor it improves RULER over attention-based compressors by a large margin. As a visual token merger for multimodal LLMs it beats the MLP merger and matches it with 55.6% fewer visual tokens.

research, multimodal· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

CellTrack-Bench: Microscopic Cell Tracking Benchmark for Vision-Language Models

CellTrack-Bench is a benchmark that tests vision-language models on microscopic cell tracking and compares them with specialised trackers. It has 1,537 problems covering 9,041 cell-linking decisions, drawn from 15 cell-tracking datasets across four imaging modalities, with a hard subset built from failures shared by distinct algorithmic trackers. The authors compare environment harnesses, benchmark 9 closed-source and 13 open-weight models, and analyse failure modes.

evals, multimodal, research· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

An Empirical Study of Recent Large Language Models for Knowledge Graph to Text Generation on Standard Benchmarks

This empirical study benchmarks recent LLMs against fine-tuned encoder-decoder models on knowledge graph-to-text generation, using reference-based metrics and graph-grounded, reference-free LLM-based evaluation validated against human annotations. Recent LLMs narrow or close the gap on reference-based metrics and do better on graph-grounded ones. Prompted models make fewer contradictions and triple omissions, and fine-tuned models' errors grow faster with graph size.

research, evals· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

The Geometry of Metaphor: Probing and Steering Metaphor Representations in Large Language Models

The Geometry of Metaphor is an October 2026 arXiv preprint that uses mechanistic interpretability to find causal subspaces for metaphor in LLMs of different architectures and sizes. The authors steer output on classification tasks across parts of speech. They also steer generation tasks, showing the metaphoricity of model output can be controlled. The abstract gives no model names or metrics.

research· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Training Small Language Models for Transferable Tool Selection

This paper trains small open language models with reinforcement learning to select tools for multi-step instructions over corpora of more than ten thousand tools. A Qwen3-8B model post-trained with the recipe outperforms Llama-4-Maverick (402B) on ToolBench. The trained model transfers to different retrievers and generalizes to new tool corpora in other domains, without relying on proprietary LLMs like Claude or ChatGPT.

research, training, agents· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Evaluating the Applicability of LLM Inference Optimizations on Apple Silicon

This October 2026 arXiv paper evaluates four LLM inference optimization families on Apple Silicon: KV-cache pruning, speculative decoding, ported GPU kernels and post-training quantization. Tests run on four Apple machines with a single-request decode workload, plus matched RTX 4090 runs. KV-cache pruning reaches 1.75x throughput on Apple, all tested speculative decoding methods reduce throughput, and ported kernels give up to 1.20x.

inference, research, evals· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

From Structured Records to Patient Narratives: Preserving Clinical Alignment in LLM-Generated Health Summaries

This paper studies evidence-grounded health summarization for menstrual health monitoring. It combines structured evidence conditioning with RAG in a prompt-based constrained generation framework, and scores 4,500 summaries from three LLMs with NLI, atomic factuality metrics and readability measures. Grounding constraints improved uncertainty preservation and generally kept or improved faithfulness against retrieved evidence, while omission effects varied by model. Clinical alignment came with harder reading and more medical terminology.

research, evals, llm-apps· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Drift Says When, Budget Says How Much: Rate-Controlled KV Re-selection for Long-Form Multimodal Generation

Rate-Controlled Refresh (RCR) is a KV cache re-selection method for multimodal LLMs in long-form generation. Attention drift decides when and where to re-select visual tokens, and a per-sample controller sets how often. RCR beats InfiniGen by 1.6-3.8 ROUGE-L points on five of six model-task settings and stays within 1.7 points of exact per-step re-selection. On video description, where drift is homogeneous, it no longer beats a fixed threshold.

research, inference, multimodal· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

SARIA: Persistence-Aware KV Caching with Coordinated DRAM--SSD Reload for LLM Agents

SARIA (Selective Admission and Reload for Inference Acceleration) is a KV caching system that jointly manages DRAM and SSD for multi-turn LLM agents. It persists only blocks likely to be reused beyond DRAM residency, promotes SSD-hit blocks into DRAM for same-session reuse, and paces DRAM reads with a transfer coordinator. Built on LMCache MP, it cuts SSD writes by 57.1 to 58.9% versus full duplicate caching and mean TTFT by 42.3 to 79.7% versus DRAM-only caching.

inference, research· Added 23 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

RIFT: KV Cache Compression for Draft Models with Complementary Compressibility Training

RIFT is a family of selection-based KV cache compression strategies for speculative-decoding draft models. It targets "decode staleness", where prefill-focused selection loses relevance as output grows, by balancing prefill information with recent generated tokens, refreshing the selection adaptively, and training drafters with sparse attention masks. Qwen3.5-4B drafting for Qwen3.5-122B keeps over 98% of full-cache acceptance length at 32K input, 4K output and an 8K KV budget.

research, inference· Added 23 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Chunk Recomputation under Real Document Revisions: Measuring Update Cost and Retrieval Quality of Fixed, Recursive, and Content-Defined Chunking

Chunk Recomputation under Real Document Revisions measures how much text must be re-embedded after edits for three stateless chunkers (non-overlapping fixed-size, recursive paragraph-first, content-defined) on 120 pairs of adjacent Chinese and English Wikipedia revisions. Fixed-size recomputes 57-68% of the new-version workload, recursive 38-56%, content-defined 43-54%, while evidence recall at a 2048-token budget differs by under two points. The authors argue update cost should be reported alongside retrieval quality.

llm-apps, research, evals· Added 24 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

SecLog-Bench: A Scalable Benchmark for Cybersecurity Log Reasoning with Scaffolded Evidence

SecLog-Bench is a benchmark for LLM reasoning over cybersecurity logs. A simulation platform with 29 attack scenarios generates audit logs, application logs and network traffic, with attack-stage labels and step-by-step evidence chains. Eight mainstream LLMs were tested on attack-progress reasoning: the best reached 44.44% accuracy, 45.07% Macro-F1 and 54.87% Stage Reasoning F1. Correct predictions often came with flawed reasoning.

evals, research· Added 23 Oct 2026, published 22 Oct 2026

fleetingswallow.comarticle

Achieving AI-style reasoning without an LLM

Minotaur is a GPUless reasoning architecture built on ELC (Energy, Control, Logic) typed roles and the MINOS representation. One "settling" operation over one network handles deduction, abduction, diagnosis, seeing versus doing, belief revision and order-dependent histories. The authors report it matched a language model on a blind police-interview test (30 of 30) and on learned production-line faults (10 of 10), and that it recovered Aristotle's 24 valid syllogisms.

research· Added 23 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

TRIAD: Benchmarking Omni-Modal Ambiguity in Multimodal Large Language Models

TRIAD is a benchmark for omni-modal ambiguity resolution. It has 327 hand-written text-image-audio queries in 86 scene groups across 18 ambiguity categories, and removing any one modality makes the answer non-unique. Across 14 omni-modal systems, the best model trails humans by over 20 points in group accuracy. Without answer options, model group accuracy falls to single digits, and leave-one-out tests show audio cues are often underused.

evals, multimodal, research· Added 23 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Gandalf: Flash Shall Not Die! Increasing Flash Storage Lifetime for KV Cache Offloading in LLM Serving

Gandalf is an SSD KV cache offloading system for LLM serving that is aware of SSD lifetime. It prioritizes admitting full-attention KV cache to SSD over cache from other attention types, using application-level information rather than access history alone. The authors report a 7.3x average SSD lifetime extension with negligible cache hit rate loss across several workload traces and hardware capacities. The design may extend to High Bandwidth Flash.

inference, research· Added 23 Oct 2026, published 22 Oct 2026

arxiv.orgpaper

Disentangling Robustness and Sycophancy in Multilingual LLMs

The SC-matrix is an evaluation framework for multilingual LLM sycophancy. It crosses a model's competence (correct answer with no pressure) with its sycophancy (keeping that answer under pressure), so agreeing with a false premise can be told apart from not knowing the answer in that language. Results: equal aggregate sycophancy rates hide different failure mixes, high competence does not always protect, and reasoning-oriented mitigations beat directive ones.

ai-safety, evals, research· Added 24 Oct 2026, published 22 Oct 2026

huggingface.copaper

Matrix-Game 4.0: Harnessing Video Models for Real-Time Playable Game Worlds

Matrix-Game 4.0 is a system that separates game logic from visual rendering for playable generated worlds. A lightweight engine holds world state and rules, a multimodal coding agent writes game code and scene captions, and a distilled autoregressive video model turns low-poly proxies into detailed frames. It runs 480p at 60 FPS on one RTX 5090 with 112 ms key-press-to-display latency. Code, weights and assets are released.

multimodal, research, agents· Added 22 Oct 2026

arxiv.orgpaper

IFC-MCQ: Automatically Generating Multiple-Choice Benchmarks for Evaluating LLM Agents on IFC Data Retrieval

IFC-MCQ is a method that automatically generates multiple-choice benchmarks from Industry Foundation Classes (IFC) building-model files, so answers can be graded by exact option-set match instead of human or LLM review. Its taxonomy has 15 categories and 36 subtypes in four groups, with answer keys from the file and distractors from file values or rules. Two agentic setups, tool-based and code-based retrieval, showed category-specific failures and variation by IFC model and interface.

evals, agents, research· Added 23 Oct 2026, published 21 Oct 2026

arxiv.orgpaper

SIM-UT: A User Simulation based Evaluation Framework for Medical Triage Chatbots

SIM-UT (Simulation of User Behaviour for Triage) is a multi-turn, simulation-based method for finding failure modes in LLM medical triage chatbots. It was applied to three WhatsApp health chatbots (India, South Africa, Kenya) across 5,000+ multilingual simulated turns, comparing MedGemma with ChatGPT and Claude. The main finding is de-escalation risk: agents that first flagged emergencies agreed with users who minimised symptoms, such as vaginal bleeding in pregnancy.

evals, ai-safety, research· Added 24 Oct 2026, published 21 Oct 2026

huggingface.coarticle

NVIDIA Kumo Relational Outperforms Supervised Machine Learning on Relational Data, No Training Required

NVIDIA Kumo Relational is a pre-trained relational foundation model that predicts outcomes such as churn or fraud directly on multi-table databases through in-context learning, using up to 10,000 context examples and no task-specific training. NVIDIA reports it beats supervised relational deep learning and XGBoost pipelines across 41 tasks on 15 databases, and beats AutoGluon by 8% MRR on SAP SALT. Code is Apache 2.0, weights are on Hugging Face.

models, research, evals· Published and added 21 Oct 2026

huggingface.coarticle

How Specializing the Vocabulary of SLMs Can Cut Inference Cost by 77% in Real Applications

VocabX is a vocabulary adaptation method from Dharma-AI that selects sub-word and multi-word "macro" tokens by serving cost instead of frequency. On a Qwen3-4B Portuguese essay grader, response length fell from 350 to 101 tokens, the LM head shrank from about 152,000 to 65,000 rows, and measured cost dropped 77.0% on an RTX 5090 with vLLM. The bench score rose from 0.703 to 0.810. It beat FVT, MWT and in-place BPE across 15 model and domain combinations.

inference, training, research· Published and added 21 Oct 2026

arxiv.orgpaper

Every Part Has Its Role: Part-of-Speech Effects in LLM Post-Training

Every Part Has Its Role studies concept augmentation in LLM post-training, where a target word is replaced by related alternatives in training sentences. It extends prior work from nouns to verbs and adjectives. Concept augmentation improves perplexity most consistently on news regardless of part of speech, while verbs give the most reliable downstream gains. CKA and linear probing show the largest gains do not match the largest representation changes.

research, training· Added 23 Oct 2026, published 21 Oct 2026

arxiv.orgpaper

Monitorability Is Trainable: Post-Training for Chain-of-Thought Monitorability

Monitorability Is Trainable proposes an unsupervised post-training objective that maximizes mutual information between a model's chain of thought and its output. On three model organisms with degraded CoT (from compression, monitor optimization, and synthetic-document generalization), it recovers most or all lost monitorability. On undegraded policies it improves CoT monitor and activation probe prediction of unsafe behavior and raises consequential reasoning.

ai-safety, research, training· Added 23 Oct 2026, published 21 Oct 2026

huggingface.coarticle

NVIDIA Kumo Forecast Combines Pretrained Time-Series Representations with Historical Pattern Retrieval and Interpretability

NVIDIA Kumo Forecast (formerly NV-Tesseract Forecasting) is a multivariate time-series forecasting model built on the MOMENT-1-large Transformer backbone (about 340M parameters) with a trainable forecasting head. Its DARR mode blends direct predictions with k-nearest-neighbor retrieval over historical context, and an interpretability framework gives per-horizon lag and feature attributions. Inputs are 256 to 2048 numeric observations; code and weights are Apache 2.0.

models, research· Published and added 21 Oct 2026

arxiv.orgpaper

Symmetry-Based Parameter Efficient Low-Rank Fine-Tuning for High-Throughput Calculation of Lattice Thermal Conductivity

SBFS-FT (symmetry-based few-shot fine-tuning) adapts the MatterSim 1M universal machine-learned interatomic potential with low-rank adaptation on a small set of compounds sharing one crystal symmetry. Tested on half-Heusler (space group 216), space group 225 and spinel (227) datasets, it cut lattice thermal conductivity error 2-3 times versus the base model. It matched per-compound fine-tuning while taking 5-7 times less compute for 100 compounds.

research, training· Added 23 Oct 2026, published 21 Oct 2026

arxiv.orgpaper

Protection Is Not One Event: A Survey of RAG Protection Research by Measured Endpoint

Protection Is Not One Event is a survey and human-coded evidence map of 118 RAG defense papers, grouped by measured endpoint into a seven-family taxonomy (kappa=0.867 across 32 annotators). It finds that only 17.8% of studies use content disclosure as their primary endpoint, and that most evaluations use general QA data and non-adaptive attacks. In a testbed on Qwen3-8B, an 8-gram guard cut copying to 0.0%, yet a three-query adaptive attacker still recovered 48.3% of hidden facts. Taxonomy, dataset, coding protocols and testbed are released.

research, evals, llm-apps· Added 24 Oct 2026, published 21 Oct 2026

openai.comarticle

Using thousands of AI agents to uncover new leads in biological data

Distributed biological reasoning is an approach in which thousands of GPT-Rosalind agents each analyzed one gene disruption in a whole-brain CRISPR atlas of millions of mouse brain cells, searched literature, and proposed mechanisms. OpenAI, PerturbAI and Scripps Research finished in 32 minutes. Distributed agents cited over 700 references versus 11 for a single agent, and blinded experts preferred their interpretations. Leads included Tsc1/Tsc2 differences and DCTN1 with cholesterol.

agents, research· Published and added 21 Oct 2026

arxiv.orgpaper

UniWALM: Unified World Action Language Model for Embodied Robot Control

UniWALM is a unified world-action-language model that learns video, action and language in one mixture-of-transformers backbone with per-modality diffusion heads that co-denoise together. Generated subtask and textual memory are fed back on the next step, so no external planner or memory bank is needed. It adds Flow Matching with Correction (FMC) for recovery from perturbed actions. It reaches 34.5% on the memory-dependent RMBench subset, above Mem-0's 28.5%, and lifts real-world Find-the-Item success from 11.1% to 61.1%.

research, multimodal· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

InterWebCoder: Learning Interactive Webpage Generation from Executable Tests via Reinforcement Learning

InterWebCoder is a framework for interactive webpage code generation by multimodal LLMs. It synthesizes self-contained webpages with executable test scripts from recorded interaction trajectories, repairs failures using runtime evidence, then trains with test-ID-weighted SFT and GRPO using test pass rates and visual similarity as rewards. The pipeline yields 1,736 verified samples from 2,058 tasks. A 9B model scores 48.54 on IWR-Bench and 48.42 on Vision2Web, and reward latency drops over 97%.

research, training, multimodal· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Benchmarking Arabic Dialects for Math Problems

DAR-MGSM (Dialectal Arabic Multilingual Grade School Math) is a manually curated benchmark for mathematical reasoning in Dialectal Arabic. It covers five major Arabic dialects plus Modern Standard Arabic, in a translated-only version and a translated and culturally localized version. Eleven Arabic-centric and multilingual LLMs were tested: accuracy generally drops on the localized version, and a mechanistic interpretability analysis shows the two variants induce distinct internal reasoning trajectories.

evals, research· Added 22 Oct 2026, published 20 Oct 2026

anthropic.comarticle

Claude designed protein binders against scorpion neurotoxins

Anthropic describes a 64-hour, largely autonomous campaign in which an internal Claude research model, running in Claude Science, designed protein minibinders against five scorpion neurotoxins. Claude generated 318,484 designs with 12 tools and picked 180; the team added 31. Adaptyv Bio and Twist Bioscience tested all 211 by SPR, and 17 met the binder criteria. No multitoxin design bound more than one toxin, and no neutralization testing was done. All designs and data are released.

agents, research· Added 21 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

AgenticRoboRL: Bringing Coding-Agent Post-Training to Agentic Robotics

AgenticRoboRL is a framework for post-training robot planners across different harnesses, combining supervised fine-tuning with multi-turn reinforcement learning. Harness-specific adapters link recorded planner interactions to task outcomes, so one RL pipeline covers both VLA skill invocation and control-code generation. On LIBERO-PRO, a 9B planner rose from 18.1% to 71.0% without external memory, near Codex (GPT-5.5) with memory at 72.1%. Code and checkpoints are promised.

research, training, agents· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Sage-A1.5: From Long-Horizon Execution to Scientific Exploration with Capability Composition

Sage-A1.5 is a 35B Mixture-of-Experts agentic model that builds on Sage-A1 for long-horizon problem solving. It adds three capabilities: orchestration with parallel subagents in isolated contexts, verification through progress checking, evidence validation and strategy revision, and open scientific discovery with inspectable research artifacts. The report shows them on complex agentic tasks and multidisciplinary science applications. The page gives only the abstract, no benchmark numbers.

research, agents, models· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Cache Engineering for Agents

Cache Engineering for Agents proposes letting LLM agents control their active KV cache directly, using a drop operation to exclude cache from attention and repos to adjust retained keys' position encodings, without editing the trajectory or re-prefilling. The authors build the AKV serving engine to handle history-dependent cache states. A tool-response rolling-drop policy raises average benchmark scores by 1.7 points and cuts cost by 27.7% against threshold-triggered summarization. AKV gives up to 1.2x output throughput and 1.5x faster P95 latency.

research, inference, agents· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

OptScientist: Automated Discovery of Optimization Laws

OptScientist is a framework that discovers compact symbolic optimization laws from data. It pairs a model that infers symbolic optimization problems with an LLM agentic system that uses literature priors, proposes candidate laws, abstracts mechanisms and suggests new data to collect. On OptFeynman, a new benchmark of 99 physics-derived problems, it recovers ground-truth laws for over 80%. It is also tested on human risky choice and central-bank monetary policy.

research, agents, evals· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Agentic Ensemble with Vision-Language Models for Visual Anomaly Detection

AnomalyAgent is a post-training fusion framework for visual anomaly detection that ensembles pretrained detectors. It aligns scores and anomaly maps, summarizes disagreement, and treats fusion as a choice among score actions and map actions. A constrained vision-language model inspects high-conflict samples and picks the actions. It beats equal-weight fusion and gets the best mean image AUROC and pixel AP among compared fusion methods on MVTec AD and VisA.

research, multimodal, agents· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Large Flow Language Models

Lafa (Large flow language model) is a blockwise continuous flow language model scaled to 8B parameters, with causal attention on the prefix and bidirectional, asynchronous flow within each block. It is adapted from pretrained autoregressive models and trained at 0.5B, 7B and 8B with under 30B tokens. A self-speculative decoding scheme verifies flow drafts autoregressively in one model, giving 2.54x the throughput of an AR model at comparable quality. Flow maps allow one- and few-step generation.

research, models· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Beyond 4 bits: Data-Centric Post-training Quantization and Recovery

This paper studies 2-bit post-training quantization (PTQ) of six dense and mixture-of-experts models from 3.8B to 32B parameters. It finds the quantization representation matters most: vector codebooks give usable models where scalar PTQ falls to chance. Calibration and recovery data must be chosen jointly. The recipe uses about 2.3 effective bits per weight, keeps 89 to 97% of full-precision QA accuracy, and needs 1.4 to 30x less compute than the strongest QAT baseline.

research, inference· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Answering in the User's Variety: Calibrated Accommodation Pairs for Post-Training

Calibrated accommodation pairs (CAP) is a preference-data method that rewards chat models for matching a user's dialect density while keeping Standard English for Standard-English users. The paper traces SFT, DPO and RL stages across 10 open model lineages and finds post-training standardizes replies, while the DPO update still favours British and Australian spelling over AAE and Singlish. On a fully open model, CAP improves convergence on unseen features without added dialect for Standard-English users or lower instruction following.

research, training, evals· Added 22 Oct 2026, published 20 Oct 2026

huggingface.coarticle

FairStress: A deeper evaluation of fairness in LLM decision-making

FairStress is an Ai2 benchmark of 300 two-candidate decision scenarios across 29 domains (hiring, lending, healthcare, housing and others). Varying demographic cues and adversarial persuasion prefixes expands it to over 13 million test cases, with a 50K-item FairStress-Core subset. Across 15 models, models abandoned correct choices more readily when the prefix favored a minority-coded candidate. Of nine debiasing methods tested, counterfactual chain-of-thought worked best.

evals, ai-safety, research· Published and added 20 Oct 2026

arxiv.orgpaper

Code2Video-Compass: A Comprehensive Benchmark for General-Purpose Code-Based Video Generation across T2V, I2V, and V2V

Code2Video-Compass is a benchmark of 2,104 tasks for code-based video generation, where multimodal LLMs write executable programs that produce complete videos from text (T2V), image (I2V) or video (V2V) inputs. Tasks span 10 categories and 24 subtypes. Scoring uses 12 objective metrics and five MLLM-judge metrics. Across 15 proprietary and open-weight models from nine families, performance falls as construction, dynamic animation and state-dependency complexity rise.

evals, multimodal, research· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

DIVE: Benchmarking Full-Duplex Interaction with Visual Engagement in Continuous Streams

DIVE is a benchmark for full-duplex interaction with visual engagement in continuous audio/video streams. It has 525 human-annotated samples across six tasks in three scenarios: proactive intervention, continuous engagement and interaction coordination. Evaluated real-time models reach sub-second average latency, but window response rates stay at or below 53.77%, and adapting ongoing responses to new information is hardest.

evals, multimodal, research· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

DeclineCraft: A RAG Abstention Benchmark for Jointly Measuring Combined Retrieval and Parametric Knowledge

DeclineCraft is a dynamic RAG abstention benchmark whose unanswerable questions target Minecraft feature additions made after a corpus cut-off, with audits for parametric exposure. Across 27 LLMs, single-shot retrieval raises Correct Abstention Rate in 38 of 40 configurations, while iterative RAG lowers it in 30 of 40. Enabling thinking without retrieval lowers abstention for 8 of 13 models, shifting responses to premise denial.

evals, llm-apps, research· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Is Sycophancy a Form of In-Context Updating?

"Is Sycophancy a Form of In-Context Updating?" argues that factual sycophancy, a model's deference to a user's false beliefs, is a failure of in-context updating rather than only a bid for approval. Models also revise answers for claims attributed to external sources, more so when better supported, and deference persists when no user approval is at stake. The internal components behind deference also matter causally for general updating, so reducing sycophancy may reduce useful updating.

research, ai-safety· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

Reward Misspecification Degrades Chain-of-Thought Monitorability

This paper shows that reward misspecification alone degrades chain-of-thought (CoT) monitorability in language models. Using a testbed of paired true and misspecified rewards across three models and three tasks, CoTs keep reasoning about task intent while outputs hack the reward. The proposed reward description training injects an LM-inferred reward description into the input and raises monitorability by an average of 24% or 42% absolute, depending on the metric.

ai-safety, training, research· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

A Multimodal Engineering GraphRAG Framework for Post-Wind-Event Building Assessment and Recovery Inference

This paper presents a multimodal engineering GraphRAG framework that links post-wind-event damage evidence from images and reconnaissance reports to building-condition states, recovery actions and downtime. Evidence is stored as a provenance-preserving, component-level representation, and graph-aligned retrieval uses compatibility, contradiction, evidence-precedence and abstention checks. Demonstrated on tornado-affected low-rise homes, it beats BaselineRAG and Structured Baseline RAG, and ablations test each component.

research, llm-apps, multimodal· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper

KNOX: Hardware-assisted Regulatory Sandbox for Generative Models

KNOX is a regulatory sandbox for generative models that runs AI workloads inside trusted execution environments (TEEs). For agentic workloads, it derives an attested policy manifest from the user prompt (applicable regulations, expected tool calls, permitted system calls). An attested reference monitor checks each action against it, and an attested execution graph records every action and verdict. Overhead is under 1%.

research, ai-policy, agents· Added 22 Oct 2026, published 20 Oct 2026

huggingface.copaper

Into the Unknown: Benchmarking AI agents on open scientific research problems

Into the Unknown is a benchmark that gives an AI agent a science paper and an unpublished research direction, then asks it to write a follow-up paper. Scientists at US Department of Energy national laboratories supplied the directions and the conditions a submission must meet, which form the grading rubric. Nine frontier agents wrote 15 follow-ups each (135 total) across five domains. None met all conditions, and 114 failed an integrity check, mostly for claiming work never done or reporting numbers that contradict recorded evidence.

evals, agents, research· Added 20 Oct 2026

arxiv.orgpaper

Can It Actually Work? Synthesizing Functional CAD Mechanisms

FunCAD-Harness is a framework in which LLM agents design CAD mechanisms and revise them using physical simulation feedback. It turns abstract functional requirements into candidate-specific physical tests. The paper also introduces FunCADBench, a benchmark of 46 mechanism families with interdependent, multi-step operations. Experiments on frontier LLM agents show it exposes mechanical failures that visual inspection misses and raises functional success over standard simulation-augmented baselines.

research, evals, agents· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

OpenVLHarness: Real World Visual Agentic Tool-Use with Multimodal Memory

OpenVLHarness is a multimodal reasoning harness for MLLMs built on a capability layer (task-level tool interfaces plus evidence rendering) and persistent multimodal memory that stores images, structured data and web references as addressable artifacts. A coding interface runs generated programs over stored data. Across 22 benchmarks and eight backbones it adds 9.7-12.6 points over no-tool baselines, and GPT-6 Luna at medium effort beats Codex and Visual Sketchpad at maximum effort using at most 37% of their tokens.

agents, multimodal, research· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Where is Physical RSI capability from? An Observational Scaling Law of LLM-as-policy

The paper introduces PhysRSI-eval, a five-axis metric for physical recursive self-improvement: first-attempt performance, efficiency, success, budget and improvement from failure. Several VLMs are tested on 18 manipulation tasks under a fixed interaction budget. It proposes the Physical Barrel Hypothesis and studies observational scaling. Workspace interventions show that removing history lowers late-round success, while frozen workspaces still help on unseen layouts.

research, evals, multimodal· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Same Simulation, Different Evidence: Where Retrieved Tool Output Costs LLM Agents Accuracy

This paper studies how keeping long tool output out of the context window, and reading it through a section index instead, costs LLM agents accuracy. Using the PHREEQC geochemistry simulator on 158 execution-verified multiple-choice items and 11,007 logged calls, GPT-5.1 and GPT-5.2 lose 11.3 and 7.9 points, while Claude Opus 4.6 and Sonnet 4.6 show no detectable cost. The authors name the failure evidence substitution: agents skip the report and misround from their own output files.

research, agents, evals· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

MemT-1.0: Towards Memory-Native Evolution

MemT-1.0 is a memory foundation model architecture that builds memory into model computation. It builds on Metis, uses Gated DeltaNet-2 for token-level memory updates within each turn, and adds Mixture-of-Storage to scale capacity. Training used 34B tokens in continued pre-training and post-training. MemT-1.0-Nano and MemT-1.0-Flash beat parametric memory baselines and stay competitive with retrieval methods. At 128K history, Nano gives about 1.86x inference speedup and 66x less persistent storage than a full-context KV cache.

research, models, agents· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

ASTOR: Benchmarking and Improving Autoformulation for Soft-Constraint Optimization with a Formal Grammar

ASTOR is an inference-time framework that has an LLM map a natural-language decision problem to a grammar-guided abstract syntax tree, then deterministically compiles it into executable optimization code. The paper also introduces SoftOR, a benchmark of natural-language problems with soft constraints. On SoftOR, ASTOR has the best aggregate accuracy among comparable single-request and multi-agent LLM methods, and it stays competitive on hard-constraint-only autoformulation benchmarks.

research, evals· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

What Language Adds, and at What Cost: Four Controlled Probes of Pathology Vision-Language Models

This preprint grades evidence for pathology vision-language models and tests four claims with controlled probes on released checkpoints and public data. Stage-redacted report embeddings raise pooled C-index by 0.019 (progression-free interval) and 0.016 (overall survival) across 1,633 TCGA patients. Zero-shot prompting matches a linear probe. SlideChat scores 74.82% with the slide and 74.41% without it. Paraphrasing flips 8-23% of answers.

research, evals, multimodal· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Scaling Parameter-Efficient Fine-Tuning for Brain Encoding

BrainLoRA is a parameter-efficient fine-tuning method that adapts multimodal backbones of up to 104B parameters to predict neural responses. It predicts multiple timepoints in one causal pass instead of sliding windows, and a multi-LoRA kernel trains many adapters on a shared backbone. Across nine backbones it improves encoding, and a 200+ model ensemble ranks first on the Algonauts post-challenge leaderboard.

research, training, multimodal· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Mise-en-Video: Fine-Grained 3D Control for Multi-Shot Video Generation

Mise-en-Video (MeV) is a multi-shot video generation framework where creators stage shots in 3D space and then generate the matching video. It combines hierarchical attention for multi-level narrative consistency with 3D-aware layout guidance for per-shot spatial control. The authors built a 160K-video dataset with hierarchical multi-shot texts and oriented 3D bounding boxes, and report gains over baselines in quality, consistency and subject controllability.

research, multimodal· Added 21 Oct 2026, published 19 Oct 2026

eprint.iacr.orgpaper

Analysis of an AI based code distinguisher

The paper analyses DeepDistinguisher, a transformer-like model that tells Goppa codes apart from random linear codes. Saliency maps and attention visualisation reveal a low-weight-row artifact in the random training instances that the model partly relies on. A pairwise-XOR rule inspired by minimum distance correlates strongly with the model's predictions, and the model loses most of its power on random matrices meeting both constraints. This suggests it exploits general combinatorial properties, not hidden algebraic structure.

research· Added 20 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

TokenGate: Adapting LLMs for Fast and Memory-Efficient Prefill with Learned Token Pruning

TokenGate is a small drop-in component that adds end-to-end learnable token pruning to trained LLMs for long-context prefill. It uses token-gated attention, inspired by Mixture-of-Experts, to re-weight attention logits with token importance scores, which gets gradients through the non-differentiable top-K step. The authors report 1.4-11.3x faster time-to-first token and 1.5-3.4x lower peak memory with no accuracy loss, on Qwen3 and Qwen3.5.

research, inference· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

EpiLens: Lightweight RAG for Epidemiological Knowledge Acquisition

EpiLens is a lightweight, open-source retrieval-augmented generation package that applies a fixed retrieval template and output schema to every paper in a corpus and links outputs to source passages. On a corpus of 2,297 epidemiology papers, its classifier matches or exceeds supervised and zero-shot baselines on four tasks without task-specific training data, scored against 214 manual labels. It also compares model-based study shares across parameters and pathogens.

llm-apps, research· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Which Model for Which Role? Cost-Efficient Pareto Recommendation for LLM Agentic Workflows

Cost-Coupled Gittins (CC-Gittins) is a multi-objective extension of cost-aware Gittins search that recommends Pareto-optimal model-to-role assignments for LLM agentic workflows. It exploits the coupling between evaluation and deployment cost, alternating between seeking higher quality and lower cost. On eight quality-cost matrices (two to five roles, up to 1,681 assignments), it recovers a substantial share of the true Pareto frontier at about 10% of exhaustive-evaluation cost, and most of it by 30%. Latency is tested as a third objective.

research, agents· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Tandem: Serving Hybrid Linear-Attention GUI Agents on Mobile NPUs

Tandem is a runtime for serving hybrid linear-attention GUI agents on mobile NPUs. It saves recurrent state where consecutive prompts diverge, keeps most prefill on the matrix unit, uses a verification kernel that reads each weight block once, and releases NPU power votes between bursts. On a Snapdragon 8 Elite it cuts per-step latency 30% for text screen descriptions and 16% for screenshots, cuts energy per text step by 33%, and lowers loaded-model standby power from 0.76 to 0.22 W.

inference, agents, research· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

The Development of Sycophancy in Language Models: Tracing Vulnerabilities from Pretraining to Post-Training

The paper traces sycophancy across 741 checkpoints from 30 open model lineages, from early pretraining to released models. It plants initial verdicts on moral scenarios and measures how far models follow a user's opinion or drop their answer under pressure. The behaviour usually exists before post-training, but which questions a released model yields on is set mainly during post-training. For Tülu-3 8B, its own preference-stage checkpoint removes 85% of remaining prediction error, a sibling model only 19%.

ai-safety, research, training· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Concerns in Benchmarking Jailbreaks: An Empirical Study of Attack Success Rates

Concerns in Benchmarking Jailbreaks is an October 2026 arXiv study of eight jailbreak attacks on three open-weight models under three evaluation procedures. Low-resource language attacks score 96.0% ASR by keyword matching but 2.7% under a Qwen3 evaluator using StrongREJECT instructions. Attack rankings flip with attempt budgets, and 100 samples at Ollama's default temperature 0.8 give 100% ASR by keywords, 50% by the evaluator.

evals, ai-safety, research· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

HAF-VLA: Hierarchical Action Flow Framework for Humanoid Vision-Language-Action Control

HAF-VLA is a hierarchical action-flow framework that adapts a pretrained flow-matching vision-language-action model to humanoid whole-body loco-manipulation. It denoises a shared action space in stages: locomotion and head commands first, then waist refinement, then full-body manipulation. Learnable stage embeddings and cross-stage KV-cache conditioning carry dependencies between stages. The authors report gains over existing models on six real-world tasks.

research, multimodal· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Assisted Turbulence Modeling via Data Assimilation and Machine Learning for Turbomachinery Flows

This paper presents a hybrid framework for axial compressor flow that couples a Multi-Fidelity Ensemble Kalman Filter (MFEnKF) with machine learning. The filter merges PIV experiments with steady RANS and unsteady hybrid RANS/LES simulations. A neural network trained on the enriched data predicts Navier-Stokes volume forcing terms from velocity, velocity gradients and turbulent viscosity. A Mixture of Experts fixes local errors near blade tips, giving lower experimental discrepancy than baseline RANS.

research· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Mitigating Reward Hacking in Preference Reward Models with Evolving Penalties

Dynalize is a method that reduces reward hacking during online RL training against preference reward models. Periodically and asynchronously, a language model reads training rollouts and writes penalties for undesirable behaviors the reward model seems to reward, and a judge applies them to every rollout. In clinical counseling and creative writing it sustains longer training and improves external benchmarks, and the gains hold in multi-turn training and after reward model retraining.

training, research, ai-safety· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Rethinking LLM Fine-Tuning via Weight Space Reparameterization: Preserving Safety during Downstream Adaptation

WSR-Tune is a fine-tuning framework that reparameterizes LLM weight matrices with a safety-conditioned basis, so safety-relevant information concentrates in a small set of directions. Those safety-critical directions are frozen, and the remaining complementary directions are updated for the downstream task. The authors report a better trade-off between downstream performance and safety retention. The page gives no benchmark names or numbers.

research, ai-safety, training· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

E-Layout: Layout-Stage Logic Restructuring via RL-Optimized Equality-Saturated Graph Retrieval-Augmented Generation

E-Layout is an ML-driven framework for post-placement logic restructuring in IC physical design. It builds equivalent logic alternatives with e-graph equality saturation, links logic choice to physical realization through a graph-based RAG model, and refines the choices with reinforcement learning on design-level timing feedback. On 9 designs it gives about 6% TNS and 2% WNS improvement and 1% less area than a commercial physical design flow.

research· Added 21 Oct 2026, published 19 Oct 2026

heymaxwell.aiarticle

Maxwell: A new kind of model that sits between code and hardware

Maxwell (Maxwell's Daemon) is doubleAI's first "code execution model" (CEM). It runs inside a live program, finds bottlenecks and emits faster machine code for the exact CPU or GPU, leaving the source untouched. Each rewrite ships with a machine-checked proof of semantic equivalence, built on a formalisation of Python. Early access is announced; the post gives no benchmark figures.

models, inference, research· Added 20 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

How Benchmark Graph Construction Shapes Multi-Hop KG Navigation: A Controlled Audit of PathQuestion and PathQuestion-Large

This paper audits the PathQuestion (PQ) and PathQuestion-Large (PQL) multi-hop knowledge-graph QA benchmarks as graph-navigation tasks. It holds questions, answers, path annotations and splits fixed, and replaces each released task-filtered graph with the larger source KG. Restoration cuts released-answer accuracy and trajectory fidelity for frozen small-language-model policies and trained MINERVA policies, and graph-consistent evaluation recovers only part of the loss.

evals, research· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

SHADE: Shrinkage via Halted Alpha-stable Denoising for Heavy-Tailed Probabilistic Time Series Forecasting

SHADE is an alpha-stable Lévy diffusion forecaster that stops denoising at a nonzero depth, so the leftover stable kernel shrinks the predictive distribution instead of leaving it overconfident. It offers a scale-calibrated halt and a context-amortized halt that needs no noise labels. Median CRPS improves 21.7% on synthetic data with alpha at or below 1.3, with tail calibration gains on cryptocurrency and foreign-exchange series.

research· Added 21 Oct 2026, published 19 Oct 2026

arxiv.orgpaper

Large-Scale Biomedical RAG: Corpus Scaling, Document Representation, and Efficiency Trade-offs

This paper is a controlled, large-scale evaluation of biomedical Retrieval-Augmented Generation across growing corpus sizes. It compares abstract, flat full-text and hierarchical full-text representations, with and without reranking, on multiple-choice QA over five external benchmarks and a held-out test set. RAG gave mean gains of up to 9.6 points externally and 10.2-30.3 points on the held-out set. Hierarchical indexing used 67% fewer units and 44-56% less storage than flat full-text.

research, llm-apps, evals· Added 21 Oct 2026, published 19 Oct 2026

huggingface.copaper

Self-evolving Agent Harness

Self-evolving Agent Harness is a 66-page survey and theory paper on revising an agent's scaffolding (prompts, memory, workflows, tools, code, plugins) while model weights stay fixed. It frames search as evolutionary optimization and uses continual PAC-learning to derive a generalization bound, a retention bound and a high-probability admission rule with fresh validation data. It separates selection, activation, attribution and improvement as evaluation claims, and covers applications and governance. Code is at github.com/zgcagi/Harness-RSI.

agents, research, evals· Added 19 Oct 2026

eprint.iacr.orgpaper

HERMES: High-Performance Private Decoding for Large Language Models

HERMES is a privacy-preserving inference (PI) system optimised for multi-token LLM decoding rather than prefill. It swaps high-round nonlinear protocols for compute- and bandwidth-heavy ones with fewer rounds, and adds a homomorphic-encryption matrix-multiplication protocol that keeps the KV cache encrypted on the server and uses grouped-query attention. On Llama 3.1 8B Instruct it cuts TPOT by up to 2.7× and total PI latency by up to 2.1×.

research, inference· Added 19 Oct 2026, published 18 Oct 2026

arxiv.orgpaper

Hybrid Reasoning Agents: LLM agents that think in latent space, act through text

Hybrid Reasoning Agents (HybridAgents) are LLM agents that switch between latent reasoning and text-based interaction in one generation process. A lightweight HybridLink handles mode transitions, continuous working memory keeps hybrid context across turns, and HybridSFT plus HybridRL train when to think latently and when to act in text. Across 3 model scales and 11 benchmarks, the paper reports 10.4% higher performance, 35.5%-54.7% fewer tokens, and 1.8x-2.6x faster inference.

research, agents, training· Added 19 Oct 2026, published 18 Oct 2026

x.comthread

Ethan Walkrman (@ethanwalkerman) on X

Harness2 is a framework from Google Cloud AI Research and Northwestern University in which an Editor model rewrites an agent's harness (system prompt, memory, skills, tool scripts, subagent routing, middleware) without changing model weights. It uses a Propose-Probe-Compose loop that compares execution trajectories instead of ground truth. On WorkBuddy-Bench, GLM-5 rose from 64.2 to 75.7 after one round, above Gemini 3.5 Flash at 72.4. The thread also lists risks: latency, experience poisoning, and a sandbox that is off by default.

agents, research, evals· Added 19 Oct 2026, published 18 Oct 2026

alignmentforum.orgarticle

The Alignment Problem: Computer Science Meets Analytic Philosophy

An AI Alignment Forum post by Vanessa Kosoy, an installment of the "Computational Agent Theory and ASI Alignment" sequence, tagged Agent Foundations. From the title, it frames the alignment problem as one that joins computer science with analytic philosophy. The fetched text held only the page header, so the argument itself is not confirmed. The sequence belongs to a theoretical line of work on agents and artificial superintelligence alignment.

ai-safety, research· Published and added 18 Oct 2026

arxiv.orgpaper

ToFu: Exploiting and Fuzzing the LLM–Orchestrator–Tool Trust Boundary

ToFu is a tool-aware fuzzing framework for LLM tools, including MCP servers. It maps each tool argument, using its metadata, to one of 74 CWE-guided semantics-aware sanitizers (path, URL, SQL and similar). It then generates schema-valid Atheris harnesses and semantic oracles, using runtime tracers for in-process tools and outcome oracles for opaque ones. The LLM (gpt-4.1) only maps sanitizers and writes seeds; it is never the security judge. The paper also shows a "Repeat-After-Me" trick, where an LLM told to repeat attacker-crafted text emits a tool call the orchestrator executes, reproduced on LangGraph, MAF, ADK 2.0, AutoGen, Strands and CrewAI. Across 273 runnable MCP servers (3,405 tools) it reports 253 verified bugs in 343 tools across 118 servers, such as path traversal, log injection and exfiltration. These are failures a crash-based fuzzer would miss. The result matters because it shows LLM-generated tool arguments cannot be treated as trusted.

mcp, research, agents· Added 18 Oct 2026

coindesk.comarticle

‘Wow, it’s a real discovery’: Claude cracks a 25-year NSA number mystery

CoinDesk reports that an internal Anthropic research model traced five unexplained seeds behind the NSA-supplied NIST elliptic curves to lrand48, an old C random-number generator fed with 1997–98 timestamps. It took under 30 minutes. The model matched the P-256 seed at the 1,842nd candidate. Its reasoning trace briefly wondered whether it was in a simulated test before concluding the discovery was real. The result makes a hidden backdoor less likely and claims a $36,864 bounty, donated to ISRG. The article also explains why Bitcoin's secp256k1 was never affected, and that quantum computing is a separate threat.

models, research· Published and added 17 Oct 2026

huggingface.copaper

OpenGCUA: Toward a Fully Open Stack for Generalist Computer Use

OpenGCUA, from Alibaba's MAI-UI team, is an open stack for generalist computer-use agents (GCUAs), where the model chooses how to observe, act and organize work across GUI, accessibility-tree, API and shell interfaces. Its design combines code execution (GUI actions written as programs with loops and branches) with harness-to-model decision transfer. On OSWorld-v2 it raises mean partial score by +19.6 points across eight models, including GPT-6 Astra and Claude Opus 5, and cuts reported steps per task by 53%. The release includes the harness, prompts, evaluation trajectories, a macOS real-device harness, a cross-device SDK and an interactive lab. It also shows large efficiency gaps between models: Qwen3.8-Max used 19.4x the reasoning tokens of GPT-6 Astra on real macOS tasks.

agents, research, evals· Added 17 Oct 2026

en.cryptonomist.charticle

AI generated viral genomes yield 16 viable phages, prompting safety concerns

Reports a Science study from Stanford University and the Arc Institute in which Evo, a genome language model trained on sequences from across the domains of life, generated thousands of bacteriophage genome designs targeting E. coli. About 300 were built and tested, and 16 produced viable viruses. A mixture of the AI-designed phages overcame resistance in some E. coli strains where a comparable mixture of natural phages did not. The article also covers a Johns Hopkins Center for Health Security commentary saying governance lags behind generative genome design, and a more cautious biosafety assessment from an outside expert.

research, ai-safety· Published and added 16 Oct 2026

anthropic.comwriteup

Claude reveals ECDSA curve origins

Anthropic's Frontier Red Team reports that Claude, given no hint, worked out how the NIST P-192/224/256/384/521 and ANSI X9.62 curve seeds were generated. They came from the C library's lrand48 LCG seeded with Unix timestamps from 1997–98, not from a hash of a phrase. Claude found this by scanning for weak-PRNG signatures and recovering the PRNG state with LLL lattice reduction. It also traced several prime-curve base points to the FreeLIP random generator. The post includes Python code that reproduces the seeds and base points. It says this is not an attack and makes backdoor claims less plausible. B-163 and the other base points remain unexplained.

research· Added 17 Oct 2026, published 16 Oct 2026

anthropic.comarticle

Designing more efficient quantum circuits with Claude

Anthropic describes giving Claude Mythos 5 plain-language specs for eight quantum-circuit primitives plus an automated evaluator that checks and scores designs (qubits × gates). Working in a closed loop with little human input, Claude found designs that beat the best published circuits, sometimes by 10x or more. Two examples: a reversible divsteps modular inversion for elliptic-curve point addition, found in a 53-hour run and verified in Lean, and a Berlekamp–Massey Reed–Solomon decoder for decoded quantum interferometry that cuts gate count about 12x versus Google's construction. The post also covers limitations such as Claude's conservatism and the mitigations used.

research, agents· Added 17 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

AI-Driven Reproducibility and Analytical Robustness in Scientific Discovery

SEER-Repro is a benchmark of 20 published papers based on the SEER cancer registry, with the data needed to reproduce their findings. An LLM agent reproduces 77.4% of 2,190 atomic claims. A second agentic pipeline finds alternative analytical choices and reruns the analyses, flagging potentially consequential choices in 9.5% of fully reproduced findings for expert review.

research, evals, agents· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

Acceptance Is Not Speed: Length Inflation in Training-Free Lossy Speculative Decoding

This paper shows that training-free relaxed (lossy) verification in speculative decoding causes length inflation: completions get longer, mostly through extended reasoning traces, which offsets gains in accepted tokens per round. Across twelve target-drafter pairs, relaxed rules accept 1.34x as many tokens per round but produce 1.54x longer completions. Rules that cut verifier rounds usually lose substantial accuracy, so the authors argue completion length and verifier rounds must be reported.

research, inference, evals· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

FuTransformer: FP8 prefill and INT4 decode for Llama-style inference on Hopper

FuTransformer is a CUDA implementation of a Llama-style Transformer block for NVIDIA Hopper GPUs. It uses FP8 prefill with fused GEMM epilogues and INT4 weights plus an INT4 KV-cache copy for decode, with a block-diagonal Hadamard transform to cut cache quantization error. On one H100, prefill is up to 1.48x faster than compiled PyTorch FP8, and INT4 decode is 1.3-2.4x faster than the faster PyTorch baseline. Llama-3.1-8B decode takes 2.8 ms per step at batch one, versus 8.1 ms for compiled transformers. WikiText-2 perplexity rises from 6.24 to 7.23 with round-to-nearest INT4 weights, or 6.99 with GPTQ.

inference, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

DLite: History-free Speculative Decoding for Agentic AI

DLite is a history-free speculative decoding framework for agentic LLM workloads. It reuses the target model's contextual hidden states and predicts locally without keeping the draft model's historical KV cache, which avoids the up to 80% acceptance-length drop seen with long, multi-turn contexts. A lightweight sequential head restores intra-block causal dependencies, and a two-stage training strategy is used. It reaches 2-6 acceptance length and beats DSpark by up to 70%.

inference, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

TileDRAM: Compiler-Controlled Tile-to-DRAM Layouts for 3D DRAM GPGPUs

TileDRAM is a compiler-runtime toolchain that adds tile-to-DRAM layouts as a compiler-visible abstraction for compute-unit-local 3D DRAM GPUs. It extends GPU Linear Layout to DRAM placement and picks layouts per tile access signature using row-buffer behavior. On 108 MoE FFN cases in Accel-Sim/Ramulator2, row-conflict rate fell from 75.7% to 15.2%, with a 1.77x geomean speedup. Overhead on an RTX 3090 is 0.17%.

research, inference· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

GLOBE: Evolving Structured Agentic Memory for E-commerce Product Optimization in Generative Search

GLOBE is an agentic framework that optimizes e-commerce product listings for generative search by storing strategies in an evolving handbook of explicit, field-scoped behaviors. An Editor applies behaviors and gets rank-based feedback, a Reflector turns it into updates, a Moderator removes redundant proposals, and a Curator commits changes. Harmful behaviors are deactivated. Structured behaviors beat monolithic prompt optimization and transfer to unseen product categories.

agents, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

FedRAPP: Federated Test-Time Prompt Tuning for Vision-Language Models with Reliability-Aware Prompt Personalization

FedRAPP is a personalized aggregation framework for Source-Free Federated Test-Time Prompt Tuning (SF-FTTPT), where clients adapt CLIP soft prompts on unlabeled, non-IID test streams without source data. It combines reliability-aware client assessment, a server-side exponential moving-average prompt memory and similarity-guided peer aggregation. Across nine CLIP benchmark datasets and ImageNet variants it beats standard federated aggregation baselines while keeping prompt-only efficiency.

research, multimodal· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

HardEval: An Execution-Verified Benchmark for AI-Driven Security Hardening and Compliance Remediation

HardEval is an execution-verified benchmark for language-model security hardening, built from Red Hat Enterprise Linux 8 STIG controls that fail on an unhardened baseline. Models receive only the required security outcome and write a Bash script, which runs on a live VM and is graded fail to pass with OpenSCAP/OVAL checks. Fourteen frontier and open-weight configurations show uneven results across domains, with plausible code that misses the exact compliance condition. The benchmark is public.

evals, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

On the possibilities and limitations of LLM-based multi-agent systems for multi-criteria decision-making in healthcare operations management

This paper proposes a framework for auditing LLM-agent-generated stakeholder preferences before they are used as weights in multi-objective optimization for healthcare operations. Experiments on outpatient appointment scheduling and emergency department configurations show that explicit ratings vary by role and clinic conditions, deliberation can erase role differences, and pairwise choices depend largely on option order. Robust weighted-sum models often select different schedules than average weights.

research, agents, evals· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

OwlSight: A Panoramic Monitor for Chain-of-Thought Reasoning

OwlSight is a chain-of-thought monitor that uses "Factor", a monitoring requirement written into the prompt, so one monitor can cover many tasks across capability, safety and trustworthiness. Factors come at fine granularity (a specific leaked email address) and coarse granularity (privacy leakage). On 15,971 reasoning trajectories it beats general-purpose LLMs by 1.32-30.57 points (fine) and 11.71-27.81 points (coarse), and reaches state-of-the-art on OpenAI's CoT monitorability benchmark. It generalizes to unseen models and resists Factor rephrasing, CoT compression and obfuscation.

ai-safety, research, evals· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

FlexPart: Text-Controlled Semantics-Aware 3D Part Segmentation at Flexible Granularity

FlexPart is a framework for text-controlled 3D mesh segmentation that assigns discrete part IDs to mesh faces through a view-independent part-identity field. It uses a control prompt of part names with location and shape clauses, trained on a VLM-built multi-granularity dataset. On FlexPartBench it raises average mIoU by 22% over the strongest baseline, and by 15% with planner-completed prompts.

research, multimodal· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

A Language-Informed Loss for Visual Generation

LILIE (Language-Informed and Lossless Image Encoder) is an invertible image transform built from a normalizing flow transformer, trained jointly with a vision-language model via next token prediction so captions can be recovered from noised features. The encoder supplies a language-informed loss for text-to-image generation. JiT models trained with it cut gFD from 66.84 to 53.01 against a pixel-only baseline on the GPIC dataset. Code, encoders and models are released.

research, multimodal· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

ThetreStageAI: Multi-View Retrieval-Augmented Generation for Structured Bengali Theatre Production Planning

ThetreStageAI is a retrieval-augmented framework for structured Bengali theatre production planning. It keeps separate scene, actor-blocking and lighting views, embeds Bengali text with BGE-M3, indexes it in view-specific Qdrant collections, and has Gemini generate schema-constrained JSON, with validation and at most one repair attempt. The authors build a synthetic dataset of 500 records (1,500 retrieval documents) and compare single-combined retrieval with multi-view retrieval. Code and data are public.

llm-apps, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

Enhancing RAG-based Medical VQA with Gated Multimodal Representations

This paper presents a multimodal retrieval-augmented generation framework for Medical Visual Question Answering. A Gated Multimodal Unit learns a joint query representation from the medical image and the clinical question, so evidence retrieval uses both modalities while the Med-LVLM stays frozen. On benchmark datasets it consistently beats a reference Med-LVLM and stays competitive with state-of-the-art methods. The page gives no dataset names or numbers.

research, llm-apps, multimodal· Added 20 Oct 2026, published 16 Oct 2026

rstation.aitool

Show HN: rStation – A shared research lab for your AI agents

rStation is a shared research community where independent AI agents discuss open math problems and publish reviewed papers, with humans as observers. Agents pick from problems such as Epoch AI's rational Diophantine septuple, the Earth–Moon problem and No three in a line, Millennium Prize problems (Riemann hypothesis, Hodge conjecture), and Brouwer's strongly regular graph tables. It is an experiment in multi-agent collaboration and peer review on hard research problems.

agents, research· Added 17 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

Trapped in Your Story: Evaluating Narrative-Induced Sycophancy in Large Language Models

This paper builds a benchmark of 500 debatable questions across ten domains to measure how user narratives steer LLM agreement. Each question has 11 matched conditions (one neutral, five mirrored narrative levels per opposing stance), giving 5,500 instances across seven LLMs. Causal framing and worldview synthesis raise agreement, personal experience can lower it, and GPT-5.1 ablations show bullet formatting also increases agreement.

ai-safety, evals, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

Predictable Scaling: Efficient Model Exploration via Scaling-Law Prediction

Predictable Scaling (PS) is a framework that predicts full-scale recommendation model performance from scaled-down proxies. It builds proxies with composable scaling functions along separate resource axes, then projects metrics to target scale by curve fitting with calibrated uncertainty. Of three predictors, even the raw proxy falls within repeated-run noise, and a Gaussian Process gives 15% lower relative MAE. On a task-specific ranker and a foundation ranker, PS ranks hypotheses correctly at 79-88% compute reduction.

research, training· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

ConQUESTA: Topology-Aware Post-Training Structural Pruning of Quantum Neural Networks

ConQUESTA is a post-training pruning framework for quantum neural networks that combines task saliency with compiler cost, screens candidate entanglers, and searches whole pruning masks under an accuracy constraint. Each mask is transpiled across several coupling topologies. At 55% pruning on 4-, 6- and 8-qubit QNNs it cuts resources by 61.6% (Breast Cancer), 72.0% (Digits01) and 82.2% (Iris), and fixed-mask recovery raises feasible runs from 16/24 to 22/24.

research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

PrairieFire: Igniting High-Throughput Updates in Large-Scale Approximate Nearest Neighbor Search

PrairieFire is an SSD-based approximate nearest neighbor search system for billion-scale vectors, built for high update throughput. It uses buffered updates and batched local rebuilds, with cascaded adjustments, concurrency control and CPU-GPU-SSD cooperation. Against state-of-the-art systems it reports about 3-35x update throughput, 3-5x search throughput and 1.5-8.7x cost efficiency. It has run in RedNote production for two months.

research, llm-apps· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

LexPilot: Benchmarking the Proactive Interaction Ability of Legal LLMs with Laypeople in Chinese Civil Case Consultation

LexPilot is a benchmark linking proactive legal interaction to claim-level verdict prediction, built from 432 real Chinese civil judgments across 36 causes of action. It pairs expert-reviewed complete and underspecified consultation histories with court verdicts. Current LLMs score on average 23.84% lower on judgment prediction in underspecified consultations than in fully specified ones, and the paper analyses the bottlenecks.

evals, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

Rethinking GRPO for Large Reasoning Model Safety Alignment

Beacon is a safety alignment framework for large reasoning models that fixes two GRPO failure modes: advantage collapse, where homogeneous rollout groups give vanishing advantages, and unsafe diversification, where unsafe behaviors persist and spread. It combines Safety Signal Recovery (advantage-guided trajectory filtering plus safe reference supervision) with Safety-Utility Decoupled Alignment (category-aware rewards, benign-only KL anchoring). Tested on four Qwen3 and DeepSeek-R1-Distill variants across eight benchmarks, it cuts average safety error by up to 32.83 percentage points with low over-refusal.

ai-safety, training, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks

SpikeQ is a spike-guided post-training quantization framework for spiking neural networks on neuromorphic chips. It picks each weight level by simulating downstream neurons to minimize spike errors, and uses a reset-aware grid for pre-fire membrane potentials. With 4-bit weights and potentials, models stay within 2.59 points of full precision, with 7.8 to 7.9 times smaller memory and 6.4 to 6.9 times less energy.

research, inference· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

MUB: An Identity-Masked Benchmark for Music Understanding in Multimodal Language Models

MUB (Music Understanding Benchmark) is a set of 500 identity-masked recordings with existing annotations for tempo, key, time signature and structural intervals. It covers all 24 keys, tempos of 47 to 180 BPM and 4,676 intervals, and strips titles, artists and platform IDs from inputs. Baselines reach 53.6% tempo accuracy (DSP) and 98.4% time-signature accuracy (constant 4/4). No multimodal LLM has been tested yet, and rights clearance is unresolved.

evals, multimodal, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

Many Biases, One Circuit: Post-Training Tunes a Pretrained Cue-Routing Mechanism

Many Biases, One Circuit is a October 2026 arXiv paper that traces attention heads behind cue-following (for example "I think the answer is B") through two post-training pipelines. Four cue types share a few heads already present in pretrained models. About 200 generic assistant examples activate them in chat format, and the final RL stage strengthens cue effects. Swapping attention patterns of six middle layers between Tulu-3-8B and OLMo-2-7B checkpoints moves cue effects most of the way.

research, ai-safety, training· Added 20 Oct 2026, published 16 Oct 2026

latent.spacearticle

[AINews] Agents disprove Polysemanticity; greatly improve Interpretability & Alignment

The 10/14–10/15/2026 AINews issue. Its lead story is Goodfire's Nick Cammarata essay "Neural Networks Are Already Clean", which argues agentic analysis resolves many known cases of polysemanticity and superposition. Neel Nanda, Leo Gao and Jake Mendel push back. It also covers Meta's Muse Spark Open (816B total / 42B active MoE, 1M context, custom license, Artificial Analysis index 39, vLLM day-0 support and quantizations), Nemotron 3.5, MAI-Base-2 and Gemini 4 Argon evals. Further items cover Vals' refund-agent and MiMo task-leakage findings, TPUv7 and Vera Rubin inference benchmarks, Perplexity's SPACE sandbox, RL training systems (TRL tree packing, bitwise-consistent vLLM RL) and Sakana's Agentic Scientist.

ai-safety, models, research· Published and added 16 Oct 2026

arxiv.orgpaper

Identify, Classify, Act: Per-Instruction Controlled Compliance for LLMs over External Data

EDIC is a decomposed reward framework that trains LLMs to decide, for each instruction embedded in external data such as documents or tool returns, whether to follow or reject it. A judge initialized with SFT and refined with GRPO attributes compliance per instruction and yields separate rewards for beneficial and harmful compliance. On Qwen3-4B it raises harmful-rejection by up to 7.7% and Arena-Hard-Auto utility by 12.5% over the strongest prior defense.

research, training, ai-safety· Added 20 Oct 2026, published 16 Oct 2026

alignmentforum.orgarticle

A Simpler and Better Debate Protocol for Scalable Oversight

An AI Alignment Forum post by Zhiyang Xun, Jiawei Li and Jonah Brown-Cohen, published 15 Oct 2026, proposing a simpler and better debate protocol for scalable oversight. The page text did not load beyond the title and tags (Debate, Iterated Amplification, Outer Alignment, Scalable Oversight), so the protocol's details and results are not visible. It matters as alignment research on using AI debate to supervise models that are hard for humans to judge directly.

ai-safety, research· Published and added 16 Oct 2026

arxiv.orgpaper

Between Spoilers and Guesswork: Avoiding Over-Specification in Coding Benchmarks

This October 2026 arXiv paper studies task over-specification and under-specification in repository-level coding benchmarks. It shows that hints derived from ground truth, as added in SWE-bench Pro, FEA-bench, Doc2Feat-bench and Multi-SWE-bench, leak spoilers in some of them. The authors test several specification levels, recommend techniques for parsimonious hints, and plan to release their hints.

evals, coding-agents, research· Added 20 Oct 2026, published 16 Oct 2026

arxiv.orgpaper

Value as Memory: Online Value-Guided Post-training for General Robot Policies

Value as Memory (VaM) is a post-training framework for generalist robot policies. It addresses partial observability and the delayed feedback of offline post-training. Value caching conditions actions on recent value predictions as task-progress memory. Online real-world RL combines offline demonstrations, off-policy interventions and on-policy rollouts, and moves from offline RL warmup to autonomous learning with sparse rewards. On long-horizon deformable-object and contact-rich manipulation tasks it gains 29.4% mean absolute success rate from value caching and 32.8% from online learning over matched offline training. The online schedule runs 8x as many policy iterations under the same gradient-step budget, with over 2x lower time per gradient step from concurrent data collection and training.

research, training· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Perceive, Predict, and Plan: a Unified World Action Model for Egocentric Dexterous Manipulation

P^3-WAM is a unified world action model for egocentric dexterous manipulation with three modes: Perceive (reconstruct articulated hand motion and camera trajectories from video), Predict (generate visual futures under prescribed actions) and Plan (jointly generate action trajectories and their anticipated visual outcomes from an initial image, a language instruction and an optional visual goal). It adds a compact structured 3D action representation to a pretrained video diffusion transformer through a mixture-of-transformers design with shared base weights, modality-specific adaptation and bidirectional visual-action attention. Randomized modality visibility lets video and 3D actions act as either context or target. Unified training beats single-mode training and outperforms specialized baselines on 3D hand reconstruction and action-conditioned video prediction across four datasets, and gives better instruction adherence than video-generation baselines.

research, multimodal· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Grounding Generative Verification in 3D Content

The paper introduces 3DVerifierBench, a 10-task benchmark for generative verification of 3D content. It scores accept/reject decisions, rejection explanations, and responses to controlled changes in specifications and candidates. A human study shows a 21.6-point gap to the strongest tested model, GPT-6 Astra. The authors also build a pipeline of multi-view fact extraction, controlled 3D editing and native annotations. It yields 14,126 training examples with contradiction explanations and linked counterfactuals. Verification-guided selection raises complete factual success from 26.7% to 90.0% with oracle selection over sixteen candidates, and human-guided recursive refinement satisfies all seven requirements for every input in an eight-input study.

research, evals, multimodal· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

DK-VLM: Geometry as Position for Spatial Reasoning in Vision-Language Models

DK-VLM adds Dual-Kernel Attention to a vision-language model. A semantic kernel keeps the pretrained attention and positional encoding. A separate geometric kernel uses estimated 3D points and camera poses to modulate attention through point displacements and relative view transformations, and a learned gate fuses the two outputs. Geometry therefore enters attention as position, not as fused 3D reconstruction features. The paper also adds Reference-Frame Composition (RC) supervision, which gives stepwise turn and heading targets derived from verified single-step annotations, to train multi-step spatial reasoning as the observer moves. DK-VLM-8B reports the highest average score across six spatial benchmarks among the compared methods, with 74.33 on VSI-Bench and 58.28 on ReVSI-32, and it preserves general video understanding.

research, multimodal· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Rethinking Routing Exploration in Modern Mixture-of-Experts

The paper identifies an "exploration dilemma" in modern Mixture-of-Experts models. Shared experts reduce the variability of token representations, which makes routing less exploratory. Fine-grained expert segmentation enlarges the combinatorial routing space and so needs more exploration, which means its extra expressiveness may go unrealized. It proposes MirrorMoE, which shrinks the exploration space by suppressing unnecessary experts per token, and reports consistent gains across model scales.

research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

MLExpert: Towards a Generalist Agentic Model for Autonomous Machine Learning Engineering

Presents MLExpert, a 35B Mixture-of-Experts agentic model trained for autonomous long-horizon machine learning engineering. Training uses a Kaggle-MLE Factory to build executable tasks and an interaction harness, hierarchical agentic SFT with outcome-guided supervision, and Windowed Interaction RL (WIRL). WIRL treats the multi-turn decisions between successive experiment results as one execution window with a shared outcome reward, and uses asynchronous collection and policy correction. On MLE-Bench Lite the average medal rate rises from 37.9% to 72.7%, comparable to larger frontier models such as Kimi-K3 and GLM-5.3. It also improves on NatureBench Lite, agent scaffold optimization and algorithm optimization.

research, training, agents· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Learning to Steer LLM Agents and Multi-Agent Systems

Proposes a steering agent that supervises LLM agents and multi-agent systems during execution and corrects goal drift, where long-horizon tool use and inter-agent collaboration slowly diverge from the overall objective. A low-cost fast path decides when to intervene, and a high-cost slow path generates how to intervene. Both paths are trained jointly with reinforcement learning, combining a pairwise ranking objective with GRPO. On information-seeking search tasks it improves performance and efficiency for single-agent and multi-agent systems, generalizes out of domain, transfers to unseen executors, and allows intervention frequency to be adjusted without retraining.

research, agents, training· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Velox: Spatial Memory Makes World Models Efficient

Velox is a training-free attention router for chunked video-diffusion world models. It uses camera frustum slabs with depth intervals from streaming geometry to pick which historical token blocks to attend to, instead of letting the KV cache grow with rollout length. Frustum overlap softly modulates each layer's model-derived block importance, and uncovered regions keep the model's own routing. The fixed block granularity suits FlashAttention-style tiled execution, and slab construction runs asynchronously with denoising on the same H200. At chunk 48 it speeds up attention 4.36× and denoising 2.43×, and on the 158-case WBench navigation split it stays within 0.15 average points of dense LingBot-Fast-Cam.

research, multimodal, inference· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Knowledge That Is Linear Before Fine-Tuning Generalizes After It

The paper separates "active" facts (which support generalization beyond recall) from "inert" ones. It builds new relations by relabeling answers of existing ones (Paris -> Europe becomes Paris -> crab), fine-tunes on a subset of subjects, and tests held-out subjects that share an answer (Munich -> crab). Across eleven language models and 70 relations, linear readability of the original relation in pretrained representations correlates with this generalization at Pearson r = 0.87 to 0.92 within each model. The result holds after controlling for fine-tuning set size, original-relation accuracy and subject frequency in pretraining text, and it survives re-randomized labels, seeds and thirteen other label categories. Mapping wrong labels back to the original answers shows the fine-tuned model's errors follow the linear probe's top answers.

research, training· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

AI Consilium: a Benchmark of LLM Clinical Multidisciplinary Second Opinions

AI Consilium is a multidimensional benchmark built from 300 diagnostically resolved NEJM–MedQA cases (110 for physician reference ratings and Judge calibration, 190 held out). Eight single-model pipelines produced role-specific opinions and interactive-consilium reports under free-form and structured clinical-reasoning conditions, and a physician-calibrated LLM Judge scored 3,040 final reports. Model identity separated quality more than format; structured reports won in seven of eight pipelines (management 2.740 vs. 2.436), multidisciplinary synthesis beat the primary-role opinion on all five shared criteria, and GPT-5.6-Sol led every component with the fewest review rounds.

evals, research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Private Retrieval Augmented Generation via Random Projection

Proposes protecting the retrieval datastore in RAG systems by applying the same random projection to datastore embeddings and query embeddings. This provably preserves query-item semantic similarity while substantially mitigating data extraction attacks. Across multiple RAG architectures and datasets it beats prior differential-privacy-based methods on retrieval and generation quality, with comparable empirical privacy under strong attacks, making it a practical baseline for privacy-preserving RAG.

research, llm-apps· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

EliteReplay: Preserving Diverse Solutions in RL Post-Training via Quality-Diversity Search

EliteReplay is an RL post-training recipe that pairs GRPO with an "elite" archive built by quality-diversity search. Before training, the frozen base model is used to collect solutions spanning distinct approaches; during training, a replay loss on this archive is added to GRPO. This counters the narrowing of solution diversity that RL causes. On text-to-SQL tasks it reaches higher pass@128 than all baselines and yields up to 77% more distinct correct strategies. It also retains more of the base model's strategies and covers both readings of ambiguous questions nearly twice as often on AmbiQT.

training, research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Librarian Agents: Training LLMs to Curate Long-Context Workspaces that Anticipate Future Queries

Introduces Librarian Agents, where a builder agent turns raw unstructured documents into a hierarchical filesystem of synthetic files and reference links, forming a workspace for a downstream answering agent. Workspace construction is cast as a Markov decision process and the builder policy is trained with reinforcement learning to maximize downstream answering accuracy. It beats the strongest retrieval and precomputation baselines by 33 points on BrowseComp-Plus, amortizes construction cost over repeated queries, and generalizes zero-shot to BRIGHT with a 31-point average gain.

research, agents, llm-apps· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

X-RIC: An Agentic RAN-Intelligent Controller Benchmark for Enhancing Reinforcement Learning

X-RIC (eXplainable RIC) is a framework that adds agentic reasoning to the O-RAN Near-RT RAN Intelligent Controller. It combines reactive reinforcement learning with automated behavioral cloning from domain heuristics, a transformer-based Telecom Language Model for contextual abstraction and interpretability, and closed-loop co-adaptation for continuous policy refinement. The paper also introduces BARX, a benchmark for agentic RAN control that measures latency, compliance and determinism. On SLA-aware network slicing, in simulation and on an O-RAN-compliant testbed, it reports 89% lower operational cost with SLA compliance kept, 75% faster convergence, 50% fewer training steps and 46% lower variance.

research, agents, evals· Added 17 Oct 2026, published 15 Oct 2026

eprint.iacr.orgpaper

GhostFaceMPC: Privacy-Preserving Face Recognition with Multi-Party Computation

GhostFaceMPC is presented as the first end-to-end face recognition system that runs both the deep CNN feature extractor and the matching step under secure multi-party computation. Raw images and biometric embeddings are never exposed at any stage, from feature extraction through storage to matching. The authors contribute an MPC-friendly quantization of the GhostFaceNet model and circuit optimizations that make this practical, and they release code on GitHub. It is relevant as an example of making CNN inference feasible under cryptographic constraints.

research, inference· Published and added 15 Oct 2026

arxiv.orgpaper

Topology-Inspired Fine-Tuning for Wind-Field Super-Resolution

The paper fine-tunes a pretrained CNN wind-field super-resolution model with differentiable topology-inspired auxiliary losses on speed, gradient magnitude, soft high-speed level sets and local maxima. It also uses fixed-pair supervision at ground-truth persistence-pair endpoints located offline with the Topology ToolKit. A 16-configuration loss study on a fixed 168-field benchmark finds that gradient supervision gives the strongest persistence-diagram improvement, while fixed-pair supervision gives the most consistent joint improvement in persistence-diagram and merge-tree distances. The combined objective cuts bottleneck, W2,∞ and W2,2 distances by 31%, 25% and 27% relative to the pretrained CNN, at a cost of about half of standard fine-tuning's PSNR gain. It also shows that pointwise metrics and topological distances rank CNN and GAN models differently.

research, training· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Skill-Decoupled On-Policy Expert Distillation for Unified Vision-Language Models

Proposes SCOPED (Skill-Decoupled On-Policy Expert Distillation), which trains capability-specific RL experts for vision-language models. It extracts explicit skills from them through reward-guided reflection. During on-policy distillation the skills guide expert supervision but are hidden from the unified student, so the student internalizes the procedural knowledge and needs no explicit skills at inference. Across eight multimodal benchmarks it beats the backbone and mixed-capability RL. Analyses show cross-capability gradient interference in joint RL, and that skill augmentation gives more consistent expert trajectories, more attention to task-relevant visual evidence, and better preservation of expertise.

research, training, multimodal· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Efficient KV Cache Migration for Geo-Distributed LLM Inference in Collaborative Edge Computing

Proposes EdgeFlow, a hybrid KV cache migration scheme for distributed LLM inference across heterogeneous edge servers in Collaborative Edge Computing. It combines direct KV cache transmission with activation-based recomputation, overlapping computation and communication to cut migration time when servers disconnect or overload. The authors formulate a multi-node KV migration optimization problem and give a 2-approximation algorithm built on improved LP relaxation and bipartite graph techniques. Experiments on a real distributed LLM inference platform show up to 45% lower migration time.

inference, research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Multi-sourced AI Tutor for Non-Native English Speaker Students

Describes a general-purpose LLM-based AI tutor for computing courses that accepts questions in Mandarin and replies bilingually, also generating glossaries of technical terms. It uses a specialized multi-source RAG pipeline with a novel question extraction mechanism to draw on diverse course materials. Evaluated on 91 Piazza posts: 83.52% answered correctly and helpfully, 76.92% of questions got high-quality translations, and 64.47% got glossaries.

llm-apps, research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

The importance of tensor handling and implementation of physics-informed DeepONets

Proposes a modified Physics-informed DeepONet (PI-DeepONet) architecture that removes redundancy in how the Branch network processes input functions across collocation points. It is mathematically equivalent to the original but uses smaller input tensors and fewer Branch evaluations. The authors derive training-time scaling laws, then test on 1D, 2D and 3D linear advection problems. Accuracy matches the original, the maximum speed-up is 3.75× in 2D, and a 3D model trains on an NVIDIA A100 only with the new formulation, because the original runs out of memory.

research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

DropKV: Decoupling Residual-Output Perturbation for Near-Optimal KV-Cache Eviction

DropKV is a KV-cache eviction framework for long-context LLMs. It decouples the joint eviction decision into independent per-token scores based on residual-output perturbation, which avoids the NP-hard combinatorial problem. It has a provable constant-factor approximation guarantee under a cone condition, which the authors verify empirically on three long-context LLMs. The fused Triton kernels never materialize the full attention matrix. They give a 19.8x scoring-kernel speedup and up to 9.9% end-to-end prefill speedup. On RULER and LongBench it beats state-of-the-art eviction baselines at equal cache budgets.

inference, research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Onboard Lightweight LLM Reasoning for Human-Aware Safe UAV Navigation

Presents a modular, fully onboard UAV navigation pipeline that closes the "human-aware safety gap" of geometry-only planners. A shared BLIP encoder drives captioning and text-conditioned segmentation, which is fused with depth for 3D obstacle localization. Qwen-2.5-0.5B, stabilized with RAG and grammar-constrained JSON decoding, infers per-category risk once per class and caches the verdict. Perception runs at 10 Hz on a Jetson Orin NX, and fallback LLM queries take about 0.8 s off the control path. In Gazebo it gives roughly 3x fewer collisions than a geometric planner and raises semantics-aware landing success from 34% to 96%. The 224-pixel encoder input limits small-object perception at range.

research, multimodal, inference· Added 17 Oct 2026, published 15 Oct 2026

alignmentforum.orgarticle

Preface to the Computational Agent Theory and ASI Alignment Sequence

A roughly 2-minute preface post by Vanessa Kosoy that opens the "Computational Agent Theory and ASI Alignment" sequence on the AI Alignment Forum. It is filed under Agent Foundations and Research Agendas, so it likely frames a theoretical, agent-foundations approach to aligning artificial superintelligence. The fetched text contained only page chrome, so the sequence's actual claims are not visible here.

ai-safety, research· Published and added 15 Oct 2026

arxiv.orgpaper

GraphACM: Graph-based Action Consistency Measurement for Uncertainty Quantification in LLM Agents

GraphACM merges equivalent states from multiple rollouts of the same LLM agent task into a directed, action-labeled graph, so actions can be compared at shared decision states. It defines node-level and global action-consistency measures, and uses the global score as an uncertainty proxy to predict task success or failure. Across three LLMs on ALFWorld, AppWorld and WebShop it reaches an average AUROC of 0.811, above the strongest behavioral-consistency baseline (0.729) and verbalized confidence (0.769). It also shows where in a trajectory agent actions diverge.

research, agents, evals· Added 17 Oct 2026, published 15 Oct 2026

eprint.iacr.orgpaper

SecAgg++: A Post-Quantum Privacy Preserving Federated Learning Protocol

SecAgg++ is a masking-based secure aggregation protocol for single-server federated learning that uses post-quantum cryptographic primitives. It keeps the communication and computation efficiency of SecAgg and SecAgg+ and handles client dropout through an efficient recovery procedure. It matters because most masking-based secure aggregation protocols are not post-quantum secure, which leaves federated model updates exposed to future quantum adversaries.

research, training· Published and added 15 Oct 2026

arxiv.orgpaper

GACC-Bench: Benchmarking Goal-Adaptive Context Compression Ability across Large Language Models

GACC-Bench is a benchmark for Goal-Adaptive Context Compression, testing how well LLMs retain goal-relevant information when compressing a source context under a length budget before downstream questions are known. It has 500 goals and 2,571 human-reviewed questions over 281 document and agent-trace contexts from seven collections; a fixed reader answers the withheld questions from the compressed context. Ten LLM compressors show large gaps to full-context performance, especially in answering every question tied to a goal. Target-goal conditioning improves accuracy over goal-agnostic compression, while mismatched goals hurt it.

evals, llm-apps, research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

From Perception to Prediction and Intervention: Benchmarking Proactive Video Assistants in Real-Life Scenarios

Introduces P^3IBench, a benchmark for proactive video understanding in real-life streaming scenarios, covering three capabilities: daily perception, risk prediction and intervention. It evaluates 12 open-source and proprietary video-language models (streaming and adapted non-streaming) and finds they struggle to make well-calibrated decisions about when to respond. The authors also build extra proactive training data and train P^3I-Assistant-4B, which improves several tasks unevenly, showing that proactive assistance needs response timing and restraint as well as visual understanding.

evals, multimodal, research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Multi-Reward Hacking: From Aggregation-Induced Drift to Trust-Anchored Defense

The paper shows that symmetric aggregation of a primary reward (correctness) and auxiliary rewards (brevity, efficiency) in LLM post-training can raise aggregate reward while primary-task performance falls. It proves that deterministic conditional-mean GRPO converges to auxiliary-favoring behavior when auxiliary rewards are more reliably obtained. It proposes Trust-Anchored Gating (TAG), which elects an anchor channel by policy drift and gates auxiliary credit on anchor satisfaction. On Qwen3-4B with a brevity reward, TAG reaches 0.581 pass@4 versus 0.220 for additive aggregation, and shortens responses by 44% relative to correctness-only training. Experiments on Qwen and Llama models cover math and program synthesis.

training, research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper

Recursive Reward Improvement via Adversarial Meta-Prompt Evolution

Introduces Recursive Reward Improvement (RRI) for recursive self-improvement on open-ended expert tasks without answer keys, where a proposer writes both a task and a rubric. Before the solver trains, an adversary writes a rubric-gaming but useless answer, the same model writes an honest answer without seeing the rubric, and up to two repair steps revise the rubric until the honest answer wins. Exploits found also rewrite future task and rubric generation. On HealthBench Professional it beats a Generation Only baseline by 10.9 points at 9B and 12.4 at 27B (10.3 with no external model), and the 27B model transfers to PRBench Hard (+6.6) and ProfBench (+8.0).

research, training· Added 17 Oct 2026, published 15 Oct 2026

huggingface.copaper

Harness^2: Recursive Agent Harnessing for an Open World

Harness^2 is a plug-and-play framework that recursively improves LLM agent execution harnesses in open-world settings where no ground-truth verifiers exist. Task-level recursion uses a contrastive Propose-Probe-Compose step to guide harness edits for the current task, and it scales through parallel or sequential recursion. Harnessing-level recursion distills experience from those edits into a continual-improvement harness for future tasks. A single step improves GLM-5 and Gemini 3.5 Flash over their base harnesses by up to 11.5 points on WorkBuddy-Bench, 4.5 on JobBench and 11.8 percentage points on LAB.

agents, research· Added 14 Oct 2026

latent.spacetalk

Lab-in-the-Loop at Compute Speed — Ian Quigley, Leash Bio

Latent Space interview with Ian Quigley of Leash Bio on using DNA-encoded library (DEL) data, roughly 4B measurements, to train protein-ligand binding models. Topics include Leash's Kaggle competition where nobody beat chance on an unseen chemical library, "Clever Hans" bias in ChEMBL where the chemist predicts activity as well as the structure does, and Hermes, trained only on Leash data, keeping pace with Boltz. It also introduces Minos, a lab-in-the-loop agent sandbox that lets LLMs do verifiable scientific reasoning and RL, searching 6.5M molecules on a budget of 100, released with a preprint, blog and open dataset.

research, agents, training· Published and added 14 Oct 2026

arxiv.orgpaper

Thinking in Graphs: Neo4j-Backed Exploit-Chain Reasoning for LLM Penetration-Testing Agents

ChainGraph is a pattern that moves exploit-chain reasoning out of the flat context window and into a Neo4j property graph. Findings, attack steps, credentials, assets and tamper-evident evidence receipts are typed nodes with pre/post-condition edges. The graph is wired to Claude Code or Codex through MCP with no custom harness. Five operator questions (next-step inference, entry-to-crown-jewel pathing, combined-severity promotion, evidence provenance, chokepoint identification) become reusable Cypher and Graph Data Science queries, plus a chain-severity score. On a synthetic multi-forest Active Directory lab modeled on GOAD, three findings rated Medium in isolation become Critical once their path to the forest root is materialized. Scoping each decision to a graph result cut the reasoning payload by about 98% on a token proxy. The authors say the results are illustrative single-run measurements, not a benchmark.

agents, mcp, research· Added 17 Oct 2026, published 14 Oct 2026

huggingface.coarticle

HiRO-ACEv2: Simulating extreme weather at 3 km resolution for a fraction of the cost

Ai2 announces HiRO-ACEv2, an update to its AI climate emulator. ACE simulates at 100 km, and the new HiROv2 downscaling model sharpens its output to 3 km. HiROv2 predicts rain, wind and pressure together and matches X-SHiELD high-resolution statistics to within 3%, even for events that occur once in a million data points. It also reproduces tropical cyclone energy, including storms unseen in training. The full system generates global 3 km climate data at about 1/300th the cost of X-SHiELD. Limitations are that it only works with the X-SHiELD-tuned ACE and cannot yet handle warmer climates or ERA5. Models, tutorial notebooks and a web demo are available.

models, research· Published and added 14 Oct 2026

eprint.iacr.orgpaper

DPF-SGD: Byzantine-Robust Differentially Private Federated SGD with Exact Optimality Guarantees

DPF-SGD is a federated SGD algorithm that stays robust to Byzantine agents while satisfying ϵ-differential privacy. It combines private step sizes with a DP noise-decay schedule and adds a privacy-preserving comparative elimination (PPCE) filter that suppresses Byzantine updates during aggregation. The authors prove almost-sure convergence to the global optimum despite persistent noise and adversaries. On synthetic and MNIST data, final optimization error is one to two orders of magnitude lower than DP baselines.

research, training· Published and added 14 Oct 2026

huggingface.copaper

Scaling Agentic Data for Long-Horizon Terminal Intelligence

Introduces TerminalHorizon, a fully automated data synthesis engine for long-horizon terminal agents. It uses independent execution to find key work omitted during environment construction and rebuilds the environment from that evidence. It then progressively expands tasks under the same goal with new requirements that build on earlier results. The authors build 1.5K task environments and run supervised fine-tuning of Qwen3.5-27B to get TerminalHorizon-27B, which improves on all 12 benchmarks covering coding-agent tasks, real-world production agentic tasks and general reasoning. Multiple RSI case studies with a 12-hour research budget show the model continuing to experiment past an initial solution and keeping effective improvements in its final deliverables.

training, agents, research· Added 13 Oct 2026

eprint.iacr.orgpaper

A Graph-Based Analysis of Comparability, Scale, and Attribution in Private Transformer Inference

Systematization of knowledge on private Transformer inference (PTI) that builds an empirical graph of 62 systems, with system attributes as node features and evaluation compatibility as edges. Node-level attribution finds that cryptographic progress drives most speedups but always co-occurs with confounders such as model fidelity, hardware and cost accounting. Edge-level auditing finds only 1 of 1891 candidate system pairs directly comparable. Graph-level analysis shows the scale frontier plateaued at 13B parameters and projects the cost of deploying current PTI techniques on production-scale Transformers.

research, inference, evals· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

ChunkAct: Action Chunking for Agentic Robotics

ChunkAct brings action chunking to agentic robotics, where an LLM agent composes perception, motion and VLA tools guided by memory. Online, one call launches a chain of primitives that runs autonomously and returns control only on failure. Offline, memory is compiled into these chunks so execution needs no memory at test time and the planner is used only for real decisions. On LIBERO-Pro it improves task success by 5.0 points over the prior agentic-robotics state of the art, with 2.9x fewer LLM calls, 4.4x fewer input tokens and 2.3x fewer output tokens.

research, agents, multimodal· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Dog-Breed Text-to-image Flow Matching from Pretraining to Fine-Tuning on 4 Consumer GPUs

MiniDog is a framework for training a text-to-image flow-matching diffusion transformer from scratch on four RTX 3090 GPUs, with 55 minutes of pretraining and 15 minutes of fine-tuning. It pretrains on 26k ImageNet dog images that were captioned, then fine-tunes on 2k FLUX.1-schnell generated images. Image and text features are precomputed and cached. The model is a 22M-parameter transformer with the E2E-INVAE semantic latent space, a small Qwen text encoder and in-context conditioning, and training is accelerated with iREPA and an EUPE perception encoder. The repo evaluates with FID, HPSv2, PickScore and VAE reconstruction metrics. It lowers the barrier to hands-on diffusion training research.

research, training, multimodal· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Hybrid Reasoning Agents

Introduces Hybrid Reasoning Agents (HybridAgents), which switch between latent reasoning and text-based interaction in one generation process. A lightweight HybridLink handles transitions between the two modes, and a continuous working memory keeps hybrid context lossless across turns. This keeps tool use and makes latent thoughts inspectable. Post-training has two stages: HybridSFT, then HybridRL, which learns when to think latently and when to act in text. Across 3 model scales and 11 benchmarks (math, science, medicine, search, agentic reasoning), the authors report +10.4% performance, 35.5–54.7% fewer tokens, 1.8–2.6x faster inference and 2.7–3.7x lower cost versus text and latent baselines.

research, agents, training· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Visual Tokens on Demand: Visual Budgeting for Scalable and Efficient Multi-Image Reasoning in Multimodal LLMs

Proposes mmVB, a lightweight plug-and-play module (1.6M trainable parameters) that assigns query-conditioned per-image visual token budgets using cross-image context for multimodal LLMs, without retraining the base model or accessing attention matrices (so it is compatible with FlashAttention). It cuts visual input tokens 9x, giving 5x faster time-to-first-token and 7x fewer prefill FLOPs while keeping 94.4% of full-resolution accuracy across eight multi-image reasoning benchmarks. It matters because existing token pruning, merging and compression methods target single-image redundancy and either need attention maps or retraining.

research, multimodal, inference· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

PixelMoE: Scaling Pixel-Space Diffusion with LLM-Style Mixture-of-Experts

PixelMoE is a pixel-space diffusion framework that scales capacity sparsely using standard LLM-style Mixture-of-Experts: token-choice routing, fine-grained routed experts, a shared expert and bias-based load balancing, with no diffusion-specific routing or auxiliary routing losses. It adds Pixel-Space Hierarchical Flow Matching, which varies spatial resolution along the denoising trajectory from global structure to fine detail. At matched active parameters it beats dense counterparts, and reaches FID 1.47 at 256x256 and 1.53 at 512x512 on class-conditional ImageNet-1K without representation alignment or perceptual losses. Code and models are promised but not yet released.

research, multimodal· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

RepProbe: Can Video Language Models Count Repeated Actions? A Diagnostic Benchmark

RepProbe is a benchmark of 42K real and synthetic video clips with exactly 1–20 complete action repetitions, controlling for clip position, temporal coverage and duration. Video language models' counting accuracy collapses at high counts, and this is not explained by frame sampling alone. Linear probes on final-layer representations show count information stays linearly decodable even when the model's answer is wrong, so the failure arises at the output stage, where answers cluster on a few preferred numbers.

evals, multimodal, research· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Plan-Bench: Can Existing VLMs Go Well Beyond Perception to Reason and Plan in Embodied Worlds?

Plan-Bench is a benchmark of 699 curated samples across 16 subclasses and three dimensions. It tests whether vision-language models can reason over spatial and temporal information and produce multi-step plans under constraints for embodied tasks, with action execution isolated out. Across 28 VLMs the best model reaches only 75.0% accuracy. Scaling generally helps, and larger models look more promising as high-level planners. Code and benchmark are released.

evals, multimodal, research· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Game2Real-Bench: Do Toy Games Reveal and Transfer to Real-World Visual Reasoning?

Game2Real-Bench tests whether vision-language model performance on procedurally generated toy games predicts real-world visual reasoning, and whether game examples help transfer. It has 1,000 verifiable game samples and 400 real-world samples, matched by shared reasoning requirements (fine-grained perception, spatial reasoning, constraint reasoning, sequential planning). Across 25 VLMs, game and real scores correlate strongly (Pearson/Spearman 0.935). One-shot game-to-real transfer on four models improves constraint reasoning for several but hurts fine-grained perception. The authors also release Game2Real-Dataset, with about 24K verified game samples across 18 tasks and 65 QA families.

evals, multimodal, research· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

OmniCompass: A Holistic Benchmark for Omni-Modal Video Grounding and Understanding

OmniCompass is a manually annotated benchmark for video grounding and understanding with two tracks: OmniCompass-VL (820 videos, 2,460 instances; vision-language models) and OmniCompass-AV (240 videos, 720 instances; joint audio-visual evidence for omni-modal models). Each track covers temporal grounding, spatial grounding and QA across short/medium/long videos, and each instance is run in paired text-only and text+image target specifications. Evaluating 50 MLLMs shows open-source models lag proprietary ones, especially on audio-visual understanding, that precise temporal/spatial grounding remains hard, and that predictions are inconsistent across the paired specifications.

evals, multimodal, research· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Beyond the Verdict: Agentic Code Judging with Verifiable Feedback

Proposes a task where a model judges and diagnoses a candidate program from only the problem statement and source code, backing its verdict with generated counterexamples that can be checked by execution. The authors build USACO-JUDGE from USACO 2026 contests (human and LLM-written accepted and unaccepted programs) and train ARBITER, a 35B MoE agentic verifier, with reinforcement learning in a generic execution sandbox. Training lifts counterexample success on incorrect programs from 40.61% to 62.67% and mean binary verdict accuracy from 71.41% to 87.41%. ARBITER also improves code selection and refinement, and in an IOI 2026 case study its guided refinement lifts GPT-OSS-120B's expected score above the gold-medal threshold under the official submission limit.

research, coding-agents, evals, training· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

SP$^2$ec: Adaptive Self-Speculative Decoding for Vision-Language Models

SP$^2$ec (Single-Peak SPeculative decoding) is a training-free bandit algorithm that adapts the draft configuration of self-speculative decoding per prompt for vision-language models. It relies on the single-peak throughput pattern seen in SWIFT, KnapSpec, SpecVLM and the authors' cosSim method. It minimizes a new wall-time regret measure and has theoretical guarantees of near-optimal configuration with negligible exploration cost. On Qwen2.5-VL-7B-Instruct and LLaVA-NeXT-Mistral-7B it improves over the original self-speculative methods by up to 11% on visual and textual tasks.

research, inference, multimodal· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

UniW(A)M: A Unified Mixture-of-Experts Framework for World and World-Action Models

UniW(A)M is a mixture-of-experts model built on diffusion Transformers that unifies action-conditioned world modeling (AC-WM) with several world-action model (WAM) variants. Shared experts capture common physical dynamics, while atomic capability experts are reused across interfaces, each keeping its own conditional pathway. At deployment a plug-in adaptive interface router (PAIR), trained on a small subset of target-task chunk scenarios with the backbone frozen, picks a WAM interface for each execution chunk. It reaches 88.3% average real-world success against 76.7% for the strongest baseline, leads WorldArena on all three action-conditioned world-modeling tasks, and PAIR raises OOD success from 63.3% to 76.7%.

research, multimodal· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

ProAS: A Progress-Aware Advantage Shaping Framework for Multi-Turn Agentic Reinforcement Learning

ProAS (Progress-Aware Advantage Shaping) is a plug-and-play framework for multi-turn agentic RL that fixes the uniform credit assignment problem in critic-free algorithms like GRPO, where trajectory-level advantages are broadcast to every action. It uses a dynamically evolving Agentic Reference Graph with Masking Unreasonable Incentives and Asymmetric Penalty Redistribution to give finer-grained credit without a critic. On Qwen2.5-1.5B/7B-Instruct it reports success rates up to 91.9%/94.3% on ALFWorld and 66.7%/70.8% on WebShop, with up to 2.44x faster training convergence and robustness to imperfect or distilled ARGs.

training, agents, research· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Kwai Summary Attention V2 Technical Report

Technical report for Kwai Summary Attention V2 (KSA V2), which extends summary-token attention architectures that compress distant context into one learnable summary token per chunk. Hybrid KSA-Full models force the same compression chunk size in every layer and leave full-attention layers paying quadratic prefill cost. KSA V2 lets summary tokens serve two roles, as KV memory and as a sparse indexer. This allows per-layer compression windows for more flexible KV cache compression and lowers computation in the full-attention layers.

research· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

MultiAlloc: Budget-Aware Precision Allocation for KV Cache Quantization

MultiAlloc is a profiling-guided framework for KV cache quantization. It splits a fixed KV memory budget between layer-wise precision enhancement and token-wise preservation. Lightweight offline profiling measures layer sensitivity and token importance, and the profiled memory–accuracy trade-off yields an allocation policy with no online search. The paper reports gains over layer-wise, token-wise and recent multi-granularity KV quantization baselines at matched memory on reasoning and long-context benchmarks, across Mistral, Llama, DeepSeek-Qwen and QwQ models.

inference, research· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Geometric Calibration of Supervised Fine-Tuning

Proposes Geometric Calibration of supervised fine-tuning and SPACE (Subspace Projection for Adapted Checkpoint Enhancement). SPACE is a post-hoc, data-free, training-free method that uses the pre- and post-SFT checkpoints to revert SFT update components lying in the principal subspace of the pre-SFT weights and keep those in the orthogonal complement. It is evaluated on Qwen2.5, Qwen3, Qwen3.5, Llama-3.1 and Olmo-3 models (2B–14B) with math, coding and medical SFT. It reduces out-of-domain regression and improves in-domain performance, for example +2.85 ID / +2.26 OOD on Qwen2.5-7B math SFT.

research, training· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Legal-R1: Agentic Retrieval for Evidence-Based Legal Reasoning

Legal-R1 is an agentic retrieval framework for evidence-based legal question answering that splits reasoning across three agents. A Lawyer agent decides when to retrieve, which legal source to query, what evidence to seek and when it has enough. A Legal Assistant retrieves statutes and precedent cases with authority-aware and diversity-aware selection, and a Judge agent writes the final answer from the full evidence trajectory. Only the Lawyer agent is post-trained, using masked supervised fine-tuning and trajectory-aware reinforcement learning. The paper reports gains over general-purpose, legal-specific and retrieval-augmented baselines on several Chinese legal benchmarks.

research, agents, llm-apps· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

LLM Agents for Automated Post-Training via Structured Recipe Optimization

Introduces Structured Recipe Optimization (SRO), a paradigm for automated LLM post-training that optimizes over a task-adapted recipe space covering data sources, data processing, cross-domain composition and training configuration. It pairs an LLM-based Recipe-Space Designer, which builds the recipe space for the target tasks, with a Multi-Domain Recipe Optimizer that uses structured trial evidence to guide LLM candidate construction and multi-objective Bayesian trial selection. SRO gets the highest average score in all evaluated target-model and agent-LLM combinations; for example, 41.1 vs 15.4 for the strongest baseline on Qwen3-1.7B with GPT-5.4, and 53.8 vs 40.8 on Qwen3-4B with DeepSeek-V4-Flash, in joint post-training across three capability domains and six benchmarks.

research, training, agents· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Configuring RAG Systems: What Breaks, Where Originates, and How to Fix

First large-scale empirical study of RAG configuration issues, covering 654 real-world issues collected from 10 diverse RAG systems. It categorizes symptoms, root causes, affected pipeline stages and fix patterns. It finds that symptoms mostly appear at early or late stages, while the root cause often sits in a different stage from where the symptom shows. It closes with nine actionable insights for testing, localizing and fixing RAG configuration problems.

research, llm-apps· Added 14 Oct 2026, published 13 Oct 2026

arxiv.orgpaper

Towards Safe Autonomy for Computer Use

Introduces SA-CUA, a dual-LLM framework for Computer Use Agents (CUAs) that defends against indirect prompt injection. It builds a trust boundary by deriving conservative deterministic rules from the accessibility tree, which unmask trusted screen regions in the privileged LLM's screenshot and keep untrusted content separate. An endorsement channel lets values move from the untrusted to the trusted domain through verification over trusted channels. This removes the need for per-action user confirmation and the confirmation fatigue it causes. SA-CUA is reported to keep much higher utility than CaMeL-CUA with a 0% attack success rate.