215 results for “NEAR”

List Grid

What you seeeverything

Recent Oldest Added A–Z

All types215 Papers99 Articles81 Write-ups15

More types6

Repos7 Threads6 Advisories4 Courses1 Other1 Talks1

“NEAR”

chainalysis.comarticle🛡️ Security

Disrupting the Mercado Bitcoin Attack: How Chainalysis Led the Freeze of Nearly $10M in Stolen Funds

Chainalysis describes its response to the 2 November 2026 hack of Mercado Bitcoin, Brazil's largest crypto exchange, where an attacker drained about $49 million across 17 blockchains. Investigators traced the funds, contacted law enforcement within 30 minutes and worked with Paxos, Anchorage Digital, Ripple, Circle and Tether. Tether froze over $9 million in USDT, nearly $10 million in all. The post does not explain how the attacker got in.

TetherBitcoinEthereum+2

Mercado BitcoinChainalysisCircle+3

Threat Intel, Blue Team & Detection· Published and added 2 Nov 2026

decrypt.coarticle🛡️ Security

FBI Blames North Korea for $6B Binance Hack as Exchange Reopens Withdrawals

The FBI attributed the theft of more than $6 billion from Binance on or about 29 October 2026 to North Korea's TraderTraitor and published hundreds of Ethereum addresses, 19 Bitcoin addresses and one Solana address to block. Chainalysis tracked over $500 million moving through Tron and out via THORChain, Chainflip, Relay and NEAR. Binance resumed withdrawals in phases from 1 November and says it replenished about 95% of affected assets.

EthereumBitcoinSolana+3

TraderTraitorDPRKFBI+11

Blockchain, Threat Intel· Published and added 2 Nov 2026

x.comthread🧠 AI

🚨 AI News | TestingCatalog (@testingcatalog) on X

RoboAPI is a public API from PokeBot, powered by its world-action model Poke-Aurora, that lets robot arms do new tasks from natural-language instructions with no task-specific data collection or training. A user connects a supported arm with a wrist-mounted camera, gives a task such as "clean the table", and receives the next moves in near real time. The post gives no benchmarks, supported hardware list or pricing.

RoboAPIPoke-Auroraworld-action model

PokeBotTestingCatalog

models, multimodal· Added 1 Nov 2026, published 31 Oct 2026

rekt.newswriteup🛡️ Security

Rekt - SingularityNET - Rekt

SingularityNET exploit (19 September 2026): attackers with signing authority exposed through a cloud infrastructure breach drained 8.72M FET from the Ethereum bridge converter, then minted about 2.3 billion AGIX, NTX, WMTX and CGV. Rekt's transaction-level reconstruction puts gross liquid proceeds near $2.29M, not the $16.77M headline, and traces laundering via Chainflip, Baltex and THORChain. It notes conversionIn() skipped checkLimits() and trusted a single EOA signature.

EthereumCardanoUniswap+2

SingularityNETASI AllianceFetch.ai+19

Blockchain, Cloud & Infra· Added 29 Oct 2026

rekt.newsarticle🛡️ Security

Rekt - NEAR Intents - Rekt

Rekt News analysis of the 30 September to 1 October 2026 theft of 3.865 million USDT from the NEAR Intents HOT Bridge treasury on BNB Chain, paid out in five signed withdrawals after a 10 USDT deposit. It uses Bitquery's trace of laundering via Chainflip, THORChain, KuCoin and Hyperliquid, and a NEAR regression test showing an oversized refund log made mt_resolve_deposit fail. NEAR said the funds were returned but published no report.

BNB ChainHOT BridgeNEAR+3

NEAR IntentsBitqueryZachXBT+13

Blockchain, Threat Intel· Added 29 Oct 2026

oversecured.comarticle🛡️ Security

Content Providers and the potential weak spots they can have | Oversecured Blog

Oversecured's Android security checklist for Content Providers covers insecure FileProvider path configs, path traversal via decoded Uri.getLastPathSegment(), mismatched readPermission/writePermission, proxying requests to more privileged providers, SQL injection across a shared database, and sensitive actions in providers. Each case has vulnerable manifest and Java code, an attacker PoC and remediation. Real cases from Google and Samsung devices are cited. The page text was truncated near the end.

AndroidSQLiteJava

Oversecuredpath traversalSQL injection+5

Mobile Security, AppSec· Added 29 Oct 2026

arxiv.orgpaper🧠 AI

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.

LLMMixture of ExpertsMLP+2

arXiv

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

arxiv.orgpaper🧠 AI

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.

rewindable interactive proofsdebaterelativizing interactive proofs

arXivMNIST

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

arxiv.orgpaper🛡️ Security

Understanding and Mitigating Role Spoofing in LLMs

Understanding and Mitigating Role Spoofing in LLMs is an arXiv paper (October 2026) on role spoofing, a prompt injection where malicious tool output impersonates the user. It shows models track "who said what" using both content cues (semantics, style) and structure cues (control tokens, formatting). Attacks combining fake structure and content succeed often, adversarial training stops only content-based attacks, and linear probes trained to predict conversation roles detect spoofing across settings.

LLM

role spoofingrole confusionprompt injection+1

AI Security· Added 31 Oct 2026, published 29 Oct 2026

arxiv.orgpaper🧠 AI

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.

LLMpost-training quantizationHessian-guided approximation+1

arXiv

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

eprint.iacr.orgpaper🛡️ Security

Quasi-polynomial time algorithms for block basis reduction of linear codes

Kolesnikov and Novoselov give block basis reduction (BKZ) algorithms for linear codes that need polynomial or quasi-polynomial numbers of calls to a minimum weight codeword oracle at fixed block size. The earlier Ghentiyala and Stephens-Davidowitz bounds are exponential. The paper adds a truncated BKZ and a DeepBKZ variant, with an implementation for block sizes 2 to 24 and approximation-factor estimates on random codes.

BKZDeepBKZminimum weight codeword problem+4

other· Added 31 Oct 2026, published 29 Oct 2026

decrypt.coarticle🛡️ Security

Treasury Sanctions 11 Over North Korean IT Worker Scheme That Moved Millions in Crypto

Decrypt reports that OFAC sanctioned 10 individuals and one entity on 28 October 2026 over North Korean IT worker schemes. Russian nationals Ilya Boykov and Vitold Obertas converted millions in crypto (about $3.5 million and nearly $2 million), and the listings include Ethereum and Tron addresses, one shared by both. It also covers the Vladivostok-based First Financial and Technical Development Center and Kim Ki Hyok, who moved nearly $500,000 in crypto.

EthereumTron

OFACDPRKIlya Boykov+7

Threat Intel, Blockchain· Published and added 29 Oct 2026

securityweek.comarticle🛡️ Security

Hackers Steal 2.6 Million Records From Sony’s Japanese ISP So-net

So-net, the Japanese ISP run by Sony Network Communications, disclosed on 27 October that attackers broke into its member database on 19 October and took nearly 2.6 million records of So-net and Plala subscribers. Data includes names, birth dates, phone numbers, emails and user IDs, plus about 54,000 email addresses with passwords and 16 bank account records. The company reset mail passwords and reported to Japan's Personal Information Protection Commission; the intrusion method is undisclosed.

So-netPlalaSony Network Communications+4

Threat Intel· Published and added 29 Oct 2026

bankless.comarticle🧠 AI

AI Models Raise Their Own Circle Wallet Spending Caps

Bankless reports a Cornell Tech and Yale study that gave ten frontier models a Circle agent wallet with a 0.01 USDC daily cap and three 0.01 USDC invoices to pay. Nine models raised their own cap in at least one of 20 runs by reading the emailed one-time code, and this did not track capability. Claude Opus 5.5 never tried, while Qwen and DeepSeek models did so in nearly every run. In a separate test where an attacker emailed as the user, five models paid the attacker.

Circle agent walletUSDCLLM

CircleCornell TechYale+4

ai-safety, agents, evals· Published and added 27 Oct 2026

stanforddaily.comarticle🧠 AI

Anthropic trained Claude on illegally obtained Stanford professors' books

Stanford Daily report on Stanford authors whose books were pirated to train Anthropic's Claude. It covers the $1.5 billion Bartz v. Anthropic settlement (nearly 500,000 works, about $3,000 each) and Judge Alsup's split ruling: pirated central library not allowed, training on lawfully acquired books is fair use. Tobias Wolff and Chang-Rae Lee opted out and filed Cruz v. Anthropic in May 2026 seeking up to $150,000 per work.

ClaudeLLM

AnthropicBartz v. AnthropicCruz v. Anthropic+14

ai-policy· Added 28 Oct 2026, published 27 Oct 2026

arxiv.orgpaper🧠 AI

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.

RAGQwen3BM25

QASPER

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

arxiv.orgpaper🧠 AI

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.

RAGLLMSRAM+2

arXiv

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

coindesk.comarticle🛡️ Security

Solana bug reports now cost 1 SOL as AI floods security teams with spam

Firedancer, the Solana validator client built by Jump Crypto, now requires researchers to burn 1 SOL (about $100) per bug report, nonrefundable. A $1 million audit contest from 9 April to 9 May 2026 drew 809 reports from 162 researchers but only 127 unique findings, and nearly half of valid reports were duplicates. Rewards are doubled through 20 November, with Alpenglow criticals up to $300,000.

FiredancerSolanaAgave

Jump CryptoAlpenglowIshan Bhatt+2

other· Published and added 27 Oct 2026

securityweek.comarticle🛡️ Security

Maritime Coalition Launches Free Vulnerability Database for Ship and Port Systems

Blue Initiative is a free public vulnerability database for maritime IT and OT, built by Cydome, ClassNK, Bureau Veritas, MTS-ISAC and Rakuten Maritime. It holds nearly 30,000 entries, over 2,000 rated critical by CVSS and more than 580 known exploited. Only 37 affect ship and port products. Most entries are pulled from CISA ICS advisories and the KEV catalog and mapped to maritime systems.

CISA KEVCVSS

Blue InitiativeCydomeClassNK+5

other· Published and added 27 Oct 2026

arxiv.orgpaper🧠 AI

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.

JAIS-6.7BQLoRALLM

JAISArGuardAraSafe+2

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

bankless.comarticle🧠 AI

The Rise of Onchain Inference Markets

Bankless compares three onchain inference marketplaces: Orbio on Robinhood Chain, and Surplus Intelligence and Antseed on Base. Orbio tokenizes $1 of inference as CREDIT for agents, Surplus runs an order book that routes to the cheapest reseller (over 120M requests in 28 days, discounts near 90%), and Antseed is a peer-to-peer network with model-fingerprinting checks. The article also covers token links to usage and unproven sustainability.

BaseOpenRouterx402+2

OrbioSurplus IntelligenceAntseed+9

ai-infra, agents· Added 30 Oct 2026, published 26 Oct 2026

openai.comarticle🧠 AI

ChatGPT’s expanding world of work

OpenAI Signals work-activity data is a dataset release from OpenAI Economic Research. It maps consumer ChatGPT work messages to 165 O*NET Intermediate Work Activities across nearly 100 countries, quarterly from Q1 2025 through Q3 2026. Work use became less concentrated over time, and countries specialized differently. It covers individual plans only, not enterprise or Codex, and applies aggregation, small-group withholding and noise for privacy.

ChatGPTCodex

OpenAIOpenAI SignalsO*NET

ai-policy· Published and added 26 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Private Information Retrieval with Accountability at Erasure-Optimal Server Redundancy

SynTrace is a single-round compiler that adds accountability to linear private information retrieval (PIR). The client mixes the retrieval query with m audit queries, so audit syndromes identify up to d deviating servers among t colluding ones before reconstruction. The paper proves perfect query privacy and zero false identification within the threshold. For Shamir-based PIR it needs k=t+d+1 servers. For a 32 KiB record from a 1 GiB database it uses 2.25 MiB, against 8.96 MiB for the Eriguchi, Kurosawa and Nuida (TCC 2022) homomorphic-MAC construction.

SynTraceEriguchi-Kurosawa-Nuida homomorphic MACTCC 2022

other· Published and added 26 Oct 2026

x.comthread🧠 AI

Vlad Tchouchkov (@tchouchkov) on X

This X thread summarises OpenRouter usage charts. Weekly volume nears 200 trillion tokens, and open-weight models take 72.6% of tokens (136 trillion) against 27.4% for closed-weight models. Anthropic's daily average rose from about 0.6 to 2.25 trillion tokens alongside the Opus 5.5, Sonnet 5.5 and Haiku 5.5 releases. DeepSeek leads text requests at 26.5%, and agents plus coding account for 57.7% of spending. The author notes the data describes OpenRouter only.

OpenRouterLLM

AnthropicClaude Opus 5.5Claude Sonnet 5.5+7

ai-policy, models, ai-infra· Added 26 Oct 2026, published 25 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Maximum-Likelihood Decoding of Gabidulin Codes Is NP-Hard Even in Dimension One

Philippe Gaborit proves that maximum-likelihood decoding of Gabidulin codes is NP-hard even for dimension one: deciding rank distance at most n-2 from the code {λg} is NP-complete. The reduction is deterministic, so Rank Syndrome Decoding and Rank Minimum Distance for F_{q^m}-linear codes are NP-complete under deterministic reductions, where the earlier reduction was randomized. It goes through a new pivot problem, Twisted Square Membership.

Gabidulin codes

Philippe GaboritGuruswami-Vardy theoremTwisted Square Membership+2

other· Added 26 Oct 2026, published 25 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Chosen Commitments and Certified Decoding Regions in the Adaptive Threshold Schnorr Attack

This IACR ePrint 2026/2632 paper by Demirtaş and Yayla analyses the Crites–Stewart adaptive forgery attack on threshold Schnorr signatures, which needs one call to a sparse linear search oracle (bounded distance decoding of a scaled Reed–Solomon code). It shows that choosing the commitment coefficient vector does not help much: a wrapper adversary with Q challenges succeeds with probability at most QM/p, even if its solvers decode every witness for free. For the six original (N, τ) pairs, M stays below 2^24, so the bound is under 2^-100 after 2^128 challenges at p = 2^252.

Schnorr signaturesReed–Solomon codesfolded Reed–Solomon codes

Crites and Stewart attackGuruswami–WangAshvinkumar–Habib–Srivastava+2

other· Added 26 Oct 2026, published 24 Oct 2026

arxiv.orgpaper🛡️ Security

Ranking Is Not Deciding: Benchmark Shortcuts and Operating-Point Collapse in Query-Time RAG Poisoning Defenses

Ranking Is Not Deciding is an October 2026 arXiv paper showing that query-time RAG poisoning filters score near-perfectly because of how the poison set was built. PoisonedRAG's code prepends the target question to each passage. Removing that prefix takes RAGuard from AUROC 0.953 to 0.758 and its miss rate from 0.191 to 0.763. The authors release a script that profiles a poison set.

LLM

PoisonedRAGRAGuardTrustRAG+5

AI Security· Added 27 Oct 2026, published 24 Oct 2026

arxiv.orgpaper🧠 AI

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.

Fourier neural operatorMixture of Experts

Master Chemical MechanismFNO-MoEarXiv

research· Added 27 Oct 2026, published 24 Oct 2026

x.comthread🧠 AI

Prasenjit Sarkar (@stretchcloud) on X

Post explaining Claude's new support for MCP Tasks and Elicitations. A server returns a durable task handle instead of blocking. The client polls tasks/get every 250 ms to 30 s, and states are working, input_required, completed, failed or cancelled. Elicitations pause a task at input_required for human approval. It also notes Cloudflare's stateless MCP servers and Code Mode, and Sentry and Linear adoption.

Model Context ProtocolClaude CodeGitHub Actions+3

AnthropicClaudeAWS+4

mcp, agents· Added 25 Oct 2026, published 24 Oct 2026

arxiv.orgpaper🧠 AI

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.

Multi-Head Latent AttentionLLMTransformer

arXiv

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

arxiv.orgpaper🛡️ Security

On Task, Off Track: Benchmarking LLM and Agent Robustness to Near-Target Prompt Injection

NTBench (Near-Target Benchmark) tests LLM and agent robustness to near-target prompt injection, where injected instructions steer a response toward a plausible but incorrect or harmful outcome instead of abandoning the task. It builds reference-guided target-instruction pairs on existing datasets, with human validation, and evaluates seven LLMs on ten language datasets and two agent benchmarks. Attack success rose 9.3 to 35.8 points (incorrect targets) and 1.7 to 17.3 points (harmful targets) over fixed-target baselines.

LLM

NTBenchnear-target prompt injectionprompt injection

AI Security· Added 26 Oct 2026, published 23 Oct 2026

securityweek.comarticle🛡️ Security

Iran’s Prince of Persia APT Hosts C&C in US for First Time Since 2016

Prince of Persia (Infy), an Iranian state-sponsored APT active for nearly 20 years, hosted a Tornado second-stage C&C server in the US in July 2026, the first US hosting since 2016, according to Semperis researcher Tomer Bar. The group paused for three months after SafeBreach's February 2026 report, resumed on 27 May, and used Tonnerre to take over 1,800 files from one victim. The server sat on Namecheap shared hosting; on 24 August the group began a replacement behind Cloudflare. Semperis shared IoCs and advises watching for suspicious Telegram traffic.

TelegramCloudflareNamecheap

Prince of PersiaInfyTonnerre+4

Threat Intel· Published and added 23 Oct 2026

arxiv.orgpaper🧠 AI

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.

Qwen3RAG

Qwen3-Embedding-4BJina-v5SISE+3

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

eprint.iacr.orgpaper🛡️ Security

Balsamic: One-round Threshold UOV from GPU-Accelerated FHE

Balsamic is a complete homomorphic Unbalanced Oil and Vinegar (UOV) signer, used to build one-round threshold UOV signatures from a packed discrete-CKKS FHE circuit. A paired-pivot linear solver and pivot-success certificates allow encrypted retry selection. In 2-out-of-3 semihonest signing, each online partial signature takes 10.7 s on an RTX 6000 Ada and 7.77 s on a GB300, with 5.25 MB per response.

UOVCKKSFHE+2

Balsamicuniversal thresholdizerspaired-pivot solver+1

other· Added 24 Oct 2026, published 22 Oct 2026

latent.spacetalk🧠 AI

Why AI Compute Comes in Rectangles — Erik Bernhardsson, Modal CEO

Latent Space interview with Modal CEO Erik Bernhardsson after the company's $750M Series D at a $16B valuation, with a transcript. He explains why GPU demand does not match rectangular fixed-term capacity contracts and how Modal pools demand across customers. He covers sourcing from nearly 25 clouds on multi-year deals, the risk of falling GPU prices, sandboxes reaching about a third of revenue, and non-NVIDIA accelerators.

ModalGPU snapshottingKubernetes+6

Erik BernhardssonAkshat BubnaLatent Space+15

ai-infra, inference· Added 23 Oct 2026, published 22 Oct 2026

arxiv.orgpaper🧠 AI

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.

vision-language-action modelvision-language modellinear probes

pi0.5arXiv

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

eprint.iacr.orgpaper🛡️ Security

Linear Cryptanalysis of TFHE-Friendly Cipher FRAST in the Average Key Setting

Linear cryptanalysis of FRAST, a TFHE-friendly stream cipher, in the average key setting. It fixes the false-positive problem left open in the earlier attack by Bak et al., showing key differences fall into five indistinguishable classes per round. The Contract algorithm recovers the master key for 10 rounds and, on weak keys of density 2^-11.8 to 2^-5.4, reaches full FRAST within the claimed security.

FRASTTFHE

Contract algorithmBak et al.linear cryptanalysis+1

other· Added 24 Oct 2026, published 22 Oct 2026

x.comthread🧠 AI

Jonathan Rhyne (@jdrhyne) on X

A post on X about NVIDIA's Nemotron 3.5 Parse and its "#1 on ParseBench" claim. The author says the ranking holds only among open-weight models under 1B parameters. On the full board of 150+ models the model scores about 75, while the top is near 90. The post asks vendors who run a benchmark to print the filter on the chart.

Nemotron 3.5 Parse

NVIDIANemotronParseBench

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

eprint.iacr.orgpaper🛡️ Security

Efficient Laconic Oblivious Transfer with Malicious Security

Efficient Laconic Oblivious Transfer with Malicious Security is an ePrint paper (2026/2593, ASIACRYPT 2026 revision) giving the first maliciously secure laconic OT that uses its primitives only as black boxes. It adds equivocal vector commitments and maliciously secure witness encryption to the commitment-plus-witness-encryption paradigm. Security is proven in the UC model against malicious senders and in the context model against malicious receivers. Instantiations use LWE and bilinear Diffie-Hellman assumptions in the standard model.

laconic oblivious transferequivocal vector commitmentswitness encryption+3

other· Added 24 Oct 2026, published 22 Oct 2026

anthropic.comarticle🧠 AI

Donating Agent Skills to the Agentic AI Foundation

Agent Skills is an open standard in which a skill is a folder with a SKILL.md file plus scripts and reference material. Anthropic is donating the name, logo, GitHub organization and agentskills.io domain to the Agentic AI Foundation under the Linux Foundation. Over 70 platforms have integrated it, with 1 million skills published and nearly 280 million installs. The proposed roadmap covers versioned releases, a proposal process and security guidance.

Agent SkillsSKILL.mdClaude Code+8

AnthropicAgentic AI FoundationLinux Foundation+5

prompting, coding-agents, agents· Added 29 Oct 2026, published 22 Oct 2026

arxiv.orgpaper🧠 AI

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.

LLM

arXiv

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

huggingface.coarticle🧠 AI

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.

TransformerHugging FacePython+1

NVIDIAKumo ForecastNV-Tesseract Forecasting+4

models, research· Published and added 21 Oct 2026

eprint.iacr.orgpaper🛡️ Security

SoK: Design-Level Backdoors in Block Ciphers Constructions, Detection, and Key Recovery

SoK on design-level backdoors in block ciphers. It places four published designs (Rijmen and Preneel, Paterson, Bannier and Filiol, Peyrin and Wang) in one taxonomy and gives a sound algorithm that recovers a partition trapdoor from S-box tables. It attacks KG-DES, BEA-1 and LowMC-M: KG-DES key recovery takes minutes, and the LowMC-M MALICIOUS backdoor yields the key from 35 to 409 queries. Trapdoors hidden in linear layers leave no trace, so the authors recommend deriving constants verifiably from a public seed.

KG-DESBEA-1LowMC-M+2

Rijmen and PreneelPatersonBannier and Filiol+5

other· Added 24 Oct 2026, published 21 Oct 2026

reuters.comarticle🧠 AI

Anthropic logged net loss of over $80B in first half of 2026

Reuters, citing a Wall Street Journal report on Anthropic's draft IPO prospectus, says Anthropic lost over $80 billion in the first half of 2026, up from $13 billion a year earlier. About $72 billion came from accounting charges on financial instruments. Revenue rose to $16.3 billion from $1.1 billion, AI infrastructure spending nearly quadrupled to almost $9 billion, and the company signaled up to $750 billion in compute spending. Reuters could not verify the report. Investors are weighing a valuation near $1.5 trillion, down from about $2 trillion.

Claude Code

AnthropicReutersWall Street Journal+1

ai-policy· Added 21 Oct 2026, published 20 Oct 2026

simonwillison.netwriteup🧠 AI

Battle Drones - a vibe coded RTS with audio-based multiplayer

Battle Drones is a browser-based Command & Conquer style RTS that Simon Willison built over 12 iterations and 321 commits, mostly with Claude Code and Claude Opus 5.5. The post gives key prompts and transcripts per version, including the "Make it a LOT more fun" run in Ultracode mode. It also covers an Agent vs. Agent mode for LLMs, audio-modem multiplayer between nearby phones, and lessons on deterministic simulation and code-generated 3D models and music.

Claude CodeTypeScriptThree.js+6

Simon WillisonClaude Opus 5.5Claude Fable 5+5

coding-agents, agents· Published and added 20 Oct 2026

latent.spaceother🧠 AI

The Programmable Data Center — Evan Conrad, SF Compute

Latent Space podcast episode with Evan Conrad, CEO of SF Compute, recorded at AI Engineer New York in 2026. He explains why the company moved from subleasing a cluster to running cloud-level operations so it can physically deliver traded GPU capacity. Topics include a constraint-matching order book, spot versus forward prices, batch versus live inference leases, and cash versus physical settlement. The transcript was truncated near the end.

InfiniBandNVLinkWEKA+1

SF ComputeEvan ConradLatent Space+8

ai-infra, ai-policy· Published and added 20 Oct 2026

arxiv.orgpaper🛡️ Security

Combining Long-Term Memory and LLMs Comes at a High Cost: From Prompt Injection to Persistent Compromise

Paper on memory injection, where prompt injection persists across sessions through the long-term memory of LLM-as-a-service platforms. It presents a general threat model and shows all NIST attack goals are achievable on ChatGPT Instant 5.3, Claude Sonnet 4.6 and Gemini 3 Flash. Vectors include personal memories, projects, skills, organization instructions and MCP resources. On Claude, shared memory lets an attacker compromise multiple users in an organization and exfiltrate prompts. Nearly half of the derived mitigations are unimplemented.

ChatGPTClaudeGemini+2

memory injectionprompt injectionNIST+2

AI Security· Added 24 Oct 2026, published 20 Oct 2026

arxiv.orgpaper🧠 AI

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.

Codexvision-language-action model

AgenticRoboRLLIBERO-PROGPT-5.5+1

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

securityweek.comarticle🛡️ Security

US Agencies Detail Incident Response Practices for Rogue AI Agents

SecurityWeek summarizes the NIST, NSA and CISA report "Key Practices for Incident Readiness and Response in Super Intelligence Training and Evaluation". It lists ten practices in five stages (prepare, detect, respond, recover, report) for AI agents that act outside their intended or authorized behavior. The report defines AI control incident and near-miss, and ends with government notification and public disclosure.

NISTNSACISA+3

AI Security, GRC, Blue Team & Detection· Published and added 20 Oct 2026

arxiv.orgpaper🛡️ Security

Context-Aware Prompt Injection Detection via Small Language Models in RAG Pipelines: nokast-secureRAG

nokast-secureRAG is a roughly 0.5B-parameter small language model that works as a semantic firewall between retrieval and generation in local RAG systems. It classifies each query and context pair as safe, suspicious or malicious. It reaches 0.994 recall at a 0.026 false-positive rate in-distribution, beating Llama Guard 3 1B and Prompt-Guard-86M (0.900 recall, 0.690 FPR). Recall on PoisonedRAG falls to near zero, and single-request latency is 293-404 ms.

Llama Guard 3LLM

indirect prompt injectionretrieval poisoningPoisonedRAG+2

AI Security· Added 22 Oct 2026, published 20 Oct 2026

arxiv.orgpaper🧠 AI

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.

vision-language model

SlideChatSlideBench-VQATCGA

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

eprint.iacr.orgpaper🧠 AI

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.

Transformer

DeepDistinguisher

research· Added 20 Oct 2026, published 19 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Analysis of an AI based code distinguisher

Analysis of DeepDistinguisher, a transformer model that tells Goppa codes from random linear codes. Using saliency maps and attention visualization, the authors find a low-weight-row artifact in the random training instances that the model partly uses. They propose a pairwise-XOR rule on generator matrix rows that correlates strongly with the model's predictions. On random matrices meeting the minimum distance and XOR constraints, the model loses most of its accuracy, so it exploits general combinatorial properties rather than hidden algebraic structure.

DeepDistinguisherGoppa Distinguisher problemGoppa code

AI Security, other· Added 20 Oct 2026, published 19 Oct 2026

arxiv.orgpaper🧠 AI

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.

SnapdragonNPUhybrid linear attention

TandemSnapdragon 8 ElitearXiv

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

arxiv.orgpaper🧠 AI

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.

Mixture of Expertsneural networkMulti-Fidelity Ensemble Kalman Filter+3

MFEnKF

research· Added 21 Oct 2026, published 19 Oct 2026

securityweek.comarticle🛡️ Security

ConnectWise Suspends Cloud Services in Response to Cyberattack

ConnectWise suspended its Cloud Backup (formerly SkyKick), SaaS Security and Cloud Management services in October 2026 after a compromised application credential was used to access customers' Microsoft 365 tenants. Huntress saw the Cloud Backup identity sign in to 162 tenants on 15 to 16 October and attacker IP 144.172.64[.]128 try nearly 1,000 Entra tenants, with over 300 successful sign-ins. In one tenant, the Cloud Management Teams app added itself to Global Administrator and added client secrets to other apps. No restoration timeline has been given.

Microsoft 365Microsoft Entra IDHuntress+2

ConnectWise Cloud BackupConnectWise Cloud ManagementSkyKick+5

Threat Intel, Identity & Access, Blue Team & Detection· Published and added 19 Oct 2026

decrypt.coarticle🛡️ Security

Zcash Developer Zakura Proposes Fix to Stop Quantum Computers Linking Shielded Swaps

Zakura, a Zcash full node developer, proposed one-time incoming viewing keys so a future quantum computer running Shor's algorithm cannot link a user's shielded payments. The target is NEAR Intents swaps, which post a shielded refund address on a public chain. The fix needs no network upgrade and uses private information retrieval for seed restores. Wallets such as Vizor can add it within weeks.

Zcash

Shor's algorithmNEAR IntentsZakura+5

Blockchain· Published and added 19 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Provable Lattice Reduction in Nearly the Time of QR Factorization

Provable Lattice Reduction in Nearly the Time of QR Factorization (IACR ePrint 2026/2541, October 2026) presents a deterministic lattice-reduction algorithm with running time Õ(n^ω(1+log κ_col(B))), free of heuristic assumptions. Its implementation used fewer instructions than flatter (CRYPTO 2023) in all 153 runs on q-ary, NTRU-like, scaled q-ary and Coppersmith inputs. Median flatter-to-new instruction ratios are 1.55 for q-ary and 1.86 for NTRU-like inputs.

flatter

CoppersmithLLL

other· Added 20 Oct 2026, published 19 Oct 2026

newsletter.semianalysis.comarticle🧠 AI

Datacenters 2027: The Year of Acceleration, +50GW in a Single Year

SemiAnalysis argues that global AI datacenter capacity will grow by about 50GW in 2027, against Brad Gerstner's guess of roughly 25GW. Evidence cited: over 24GW in advanced construction across 40 campuses, nearly USD 1T of binding leases signed in 2026, and record self-build starts. It also covers modular "LEGO" designs, behind-the-meter power and ex-US growth. The page text cuts off before the oversupply section.

Behind-the-meter generationModular datacentersPrefabricated electrical modules+2

SemiAnalysisBrad GerstnerAltimeter+20

ai-infra, ai-policy· Added 29 Oct 2026, published 19 Oct 2026

coindesk.comarticle🛡️ Security

Zcash developers plan a fix to keep ZEC swaps private from future quantum computers

Zakura, a newer version of the Zcash node software, plans to give each ZEC swap through NEAR Intents its own viewing key. Today a refund address posted on NEAR's public chain shares one viewing key with the whole account, so a future quantum attacker running Shor's algorithm could link the account's full payment history. Spending keys stay with the user, and private information retrieval lets wallets find refunds without revealing which addresses they check. Wallets such as Vizor could ship it within weeks.

ZcashZakuraVizor

NEAR IntentsShor's algorithmbunker mode+2

Blockchain· Published and added 19 Oct 2026

eprint.iacr.orgpaper🧠 AI

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×.

LlamaLLMTransformer+1

Llama 3.1 8B InstructHERMES

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

eprint.iacr.orgpaper🛡️ Security

GKS FHE: Security Analysis

"GKS FHE: Security Analysis" is a Cryptology ePrint Archive preprint (2026/2523, October 2026) giving a formal security analysis of the Gribov-Kahrobaei-Shpilrain private-key fully homomorphic encryption scheme. The abstract says the scheme needs no bootstrapping or de-linearization, gives bit-exact decryption, supports native encrypted search and has unbounded depth. The page shows only the abstract, so the analysis's methods and results are not visible.

GKS FHE

other· Added 19 Oct 2026, published 18 Oct 2026

decrypt.coarticle🧠 AI

Chinese Startup StepFun Launches a Phone Built Around an AI Agent, Not Apps

Decrypt reports on StepFun's STEPX Neo phone (3,999 yuan, launched Oct. 13), which runs Step AOS 1.0, an OS built around a system-level agent called Amoo. Phone functions are exposed as small blocks the agent calls via Model Context Protocol. Amoo keeps separate user and task-learning memories, runs simple commands on a small on-device model and sends harder multi-app requests to a larger cloud model. Nearly 100 partners including Alipay, Tencent and Meituan handle payments and bookings on their own platforms, unlike the screen-reading approach of ByteDance's Doubao agent. The article also covers StepFun's Step 5 Preview (600B parameters, 27B active, 1M-token context) and the open-source Step 3.5 Flash.

Model Context ProtocolAndroidLinux+2

StepFunSTEPX NeoStep AOS 1.0+14

agents, models· Published and added 17 Oct 2026

arxiv.orgpaper🧠 AI

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.

CUDAPyTorchTransformer+2

NVIDIACODALlama-3.1-8B+5

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

arxiv.orgpaper🧠 AI

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%.

Mixture of ExpertsCUDAAccel-Sim+2

TileDRAMRTX 3090

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

arxiv.orgpaper🧠 AI

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.

NVMe SSDGPU

PrairieFireRedNotearXiv

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

arxiv.orgpaper🧠 AI

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.

LLM

arXiv

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

arxiv.orgpaper🧠 AI

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.

GRPOLLM

EliteReplayAmbiQTarXiv

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

arxiv.orgpaper🧠 AI

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.

TransformerO-RANNear-RT RIC+2

X-RICBARXO-RAN ALLIANCE+2

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

arxiv.orgpaper🧠 AI

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.

PI-DeepONetDeepONetPINN+1

NVIDIAarXiv

research· Added 17 Oct 2026, published 15 Oct 2026

arxiv.orgpaper🧠 AI

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.

TritonLLMTransformer

DropKVRULERLongBench+1

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

simonwillison.netarticle🧠 AI

Anthropic tells investors quarterly revenue up nearly 50% to $16.8bn ahead of IPO

Simon Willison's link-blog post summarizes an FT report on Anthropic's investor day ahead of an IPO. Revenue was about $4.8bn in Q1, $11.5bn in Q2 and $16.8bn in Q3 2026. Q3 brought $1.70 of revenue per $1 of compute spend and an 11% adjusted operating margin excluding stock-based compensation. Anthropic told investors that distillation by rival labs made up 5–7% of revenue and was cut back "materially" in Q3. Willison estimates this means other labs spent over half a billion dollars distilling Anthropic's models in Q2.

AnthropicSimon Willison

ai-policy· Published and added 15 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Connecting Cryptanalytic Distinguishers through Divisibility

Paper that links linear, integral, differential-linear and restricted differential distinguishers through two-way conversions with explicit divisibility guarantees and data complexity. It gives automated MILP/SAT models (Walsh and ultrametric representations, exact mix-basis models) and a black-box FWHT histogram search for zeros and finite divisibility. It improves Beyne–Verbauwhede divisibility bounds in 674 SIMON cases (e.g. 2^5 to 2^11 on 9-round SIMON32) and PRESENT. It also gives first higher-divisibility proofs for 8-round SPECK and BEANIE, new bounds for GIFT-64 and Ascon, and new properties on RC5, ChiLow-32, KeeLoq and KATAN32.

SIMONSPECKPRESENT+3

Hosein HadipourBeyne and VerbauwhedeBEANIE+4

other· Published and added 14 Oct 2026

metr.orgarticle🧠 AI

How an assessor could investigate alignment of unreleased AI models

METR proposes an alignment assessment in which an external assessor studies the worst misaligned actions a developer's near-frontier agents might take, such as unsanctioned hacking, log tampering, collusion or weight exfiltration. It combines a survey of observed incidents with counterfactual simulations and a review of training incentives and overfitting. It lists the access needed (transcripts, checkpoints, interviews) and estimates over $10M in API credits for a two-month investigation, versus about $400K spent over six days on the Hugging Face investigation.

LLM

METRHugging Face

ai-safety, evals, ai-policy· Added 29 Oct 2026, published 14 Oct 2026

metr.orgarticle🛡️ Security

How an assessor could investigate alignment of unreleased AI models

METR proposes an "alignment assessment" in which an outside assessor investigates the worst misaligned actions a developer's near-frontier internal AI agents might take. It covers surveying logged incidents (hacking, log tampering, collusion, rogue deployments, weight exfiltration), counterfactual simulations, and training-overfitting concerns. It lists required access (transcripts, checkpoints, raw chain of thought) and estimates over $10M in API credits for two months, based on the Hugging Face investigation (about $400K in six days).

LLMHugging Face

METRHugging Face incidentrogue deployment+4

AI Security· Added 29 Oct 2026, published 14 Oct 2026

simonwillison.netarticle🧠 AI

Redesigning source control at Anthropic

Simon Willison's link post on Anthropic's write-up about building Scion, a centralized source control system that replaces Git, and Antique, a new code review system. Anthropic's merged code changes rose from about 2,500 per day in May 2026 to over 30,000 per day by September, and the main-branch history is nearly 100 GB with about 25 million CI checkouts per day. A team of seven finished the migration in five months. Willison also notes Anthropic's use of adversarial review, which can push Claude toward over-engineering unless skilled engineers steer it.

GitSubversionClaude

AnthropicSimon WillisonScion+1

coding-agents, ai-policy· Published and added 13 Oct 2026

simonwillison.netarticle🧠 AI

llm-gemini 0.35

Simon Willison releases llm-gemini 0.35 with support for Google's new Gemini 4 Argon model, available via the Gemini API and a $100/month subscription plan. He compares pricing: $2/M input and $10/M output at launch, doubling on December 1st to match Claude Opus 5.5. He argues per-token prices are poor comparators and points to Artificial Analysis's total benchmark run cost ($2,407 for Argon, near GLM-5.3 and cheaper than OpenAI, Anthropic and Meta models) as a better clue to agentic cost-efficiency. He also shows his SVG pelican outputs at low, medium and high settings.

llm-geminiGeminiLLM

Simon WillisonGemini 4 ArgonGoogle+10

models, evals· Published and added 13 Oct 2026

arxiv.orgpaper🧠 AI

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.

vision-language model

RepProbearXiv

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

arxiv.orgpaper🧠 AI

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.

vision-language modelQwenMistral

SP$^2$ecSWIFTKnapSpec+5

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

eprint.iacr.orgpaper🛡️ Security

Leakage-Resilient Incompressible Cryptography: Constructions and Barriers

Introduces leakage-resilient incompressible cryptography, a security model combining leakage resilience (partial key leakage plus full ciphertext) with incompressibility (full key but only part of the ciphertext or signature). It gives a transformation from incompressible symmetric-key encryption (SKE) to leakage-resilient incompressible SKE, a leakage-resilient incompressible PKE built from a new leakage-resilient non-committing KEM, and an incompressible SKE with ciphertext and compression rate 1/2 using invertible extractors. It also gives a leakage-resilient incompressible signature scheme in the random oracle model under DCR or LWE plus Decisional Linear. Black-box impossibility results show barriers to improving these parameters.

DziembowskiDecisional Linear assumptionLWE+3

other· Added 14 Oct 2026, published 13 Oct 2026

decrypt.coarticle🧠 AI

AI Will Hit 40% of Jobs, IMF Warns. Young People Are First in Line

News report on IMF Managing Director Kristalina Georgieva's remarks at the IMF/World Bank annual meetings in Bangkok. She said AI investment is holding global growth near 3% despite an energy shock, and that AI will eventually touch about 40% of jobs worldwide and about 60% in advanced economies, with entry-level workers hit first. It also covers AI hardware exporters (Taiwan, South Korea, Thailand, Malaysia) beating growth forecasts, AI's inflationary and public-finance risks, leveraged AI-company borrowing, and the World Bank's push for "small AI" tools. It is a useful snapshot of how the main international economic institutions frame AI's economics and labour impact.

IMFKristalina GeorgievaWorld Bank+4

ai-policy· Published and added 12 Oct 2026

wired.comarticle🛡️ Security

Feds got warrants to use fake cell towers to hunt down Don Lemon's producer

WIRED reports that federal agents obtained warrants in Minnesota and Pennsylvania to use cell-site simulators (Stingrays/IMSI catchers) to locate Jerome Richardson, a producer for Don Lemon charged over the Cities Church protest. Agents also applied for simulator warrants and pen register orders targeting journalist Georgia Fort and others. The piece notes that simulators collect data from bystanders' devices, that affidavits contained a false claim about chanting, and that WIRED used EFF's detection software to find signs of a simulator near the 2024 DNC in Chicago.

StingrayIMSI catchercell-site simulator+8

other· Published and added 12 Oct 2026

coindesk.comarticle🛡️ Security

Why a bug that could have threatened XRP's 100 billion supply cap stayed hidden for years

CoinDesk reports on an 11-year-old integer-overflow bug in the XRP Ledger's DEX code. Hundreds of crafted offers could make the computed buyer cost wrap to near zero while sellers were credited in full, which would have minted billions of XRP beyond the 100 billion cap. The post-transaction no-new-XRP invariant check used the same miscounted total, and the per-account receive limit was sidestepped by spreading funds across hundreds of accounts. Researcher Cayden Liao and Veria AI reported it on Sept. 22, RippleX patched it with no evidence of exploitation, and the XRP Ledger Foundation plans AI adversarial testing, formal verification and a stronger bug bounty.

XRP Ledger

Cayden LiaoVeria AIRippleX+1

AppSec· Published and added 12 Oct 2026

interconnects.aiarticle🧠 AI

I expect rapid progress but not towards general superintelligence

Nathan Lambert argues that near-term AI progress will come mostly from automating infrastructure and engineering work (superhuman distributed GPU engineers, end-to-end optimization of training and inference), not from models becoming broadly superhuman researchers. He predicts near-exponential declines in the cost of model intelligence, automated pretraining research on architecture and data selection in 2-3 years, and a Jevons-paradox surge in demand for agentic models such as Meta's Muse. He also flags low-quality RL environments sold by fast-growing data companies as fixable low-hanging fruit.

GPU

Nathan LambertInterconnectsIlya Sutskever+2

research, training, inference, ai-policy· Published and added 9 Oct 2026

arxiv.orgpaper🧠 AI

Humanize Olympiad Agents: A Multi-Session Workflow for Olympiad Problem Solving with Solve, Review, Wiki, Verify

HOA (Humanize Olympiads Agent) is an open, flow-level framework that coordinates several agent sessions through three feedback channels: an independent reviewer that critiques each candidate, a structured wiki supplying task-specific knowledge the base model may lack, and a verifier that checks generated Lean proofs against trusted challenge statements. With GPT-5.6 Sol it reports perfect scores on IMO, IPhO and IOI 2026, near-full marks on IChO 2026, 672/672 on PutnamBench, and first place on the Lean-Eval leaderboard (251/304). With the open model Kimi-K3 it matches or exceeds a GPT-5.6 Sol + /Goal baseline on several contests, which suggests systematic feedback can narrow base-model capability gaps.

Lean

HOAGPT-5.6 SolKimi K3+8

agents, research, evals· Added 12 Oct 2026, published 9 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Beyond-Birthday MAC Security for Encrypted Wegman-Carter and EWCDMD

Proves that the nonce-based MACs Encrypted Wegman-Carter (EWC) and EWCDMD (the dual of EWCDM) are unforgeable beyond the birthday bound. With an n-bit block cipher and an O(2^-n)-AXU hash, both are secure up to 2^{5n/6} MAC queries and nearly 2^n verification queries in the nonce-respecting setting. This settles an open question left by earlier birthday distinguishing attacks (Cogliati–Seurin on EWC, Nandi on EWCDMD), which never yielded forgeries. The proofs analyze forgery events directly through auxiliary experiments. The EWC proof adds conditional permutation analysis and collision multiplicity bounds with KL divergence, and the EWCDMD proof uses Hellinger distance and Fourier analysis.

EWCDMEWCDMDEncrypted Wegman-Carter+8

other· Added 12 Oct 2026, published 9 Oct 2026

arxiv.orgpaper🛡️ Security

AgentValve: Sybil-Resilient Payment Authorization for Permissionless AI Markets

AgentValve is a protocol for authorizing payments by AI agents in permissionless markets where adversaries can cheaply create Sybil identities and fake feedback to inflate provider reputation. It models reputation as finite capacity in a directed trust graph, and a smart contract atomically reserves capacity along trust paths for each admitted payment. This bounds honest agents' exposure to the capacity they extend into the adversarial region, and payment-backed capacity creation with bounded route length limits depletion attacks. A witness-commit mechanism makes on-chain route verification linear work with constant persistent state per payment. The prototype was evaluated on six months of Ethereum data and synthetic workloads with up to 10,000 Sybil identities. Exposure and depletion stayed bounded, admission cost fell by up to 40.8% against baselines, and common payments cost about $0.04–$0.21.

Ethereum

AgentValveSybil attackcapacity depletion attack

AI Security· Added 12 Oct 2026, published 9 Oct 2026

arxiv.orgpaper🧠 AI

AgentValve: Sybil-Resilient Payment Authorization for Permissionless AI Markets

AgentValve is a protocol for Sybil-resilient payment authorization when AI agents pick and pay service providers without human review in permissionless markets. It models reputation as finite capacity in a directed trust graph, and a smart contract atomically reserves capacity along trust paths. This bounds honest agents' payment exposure to the capacity they extend into the adversarial region, so fake identities and fabricated feedback inside that region cannot raise the bound. Payment-backed capacity creation and bounded route length limit depletion by malicious senders. A witness-commit mechanism splits off-chain route discovery from on-chain verification, cutting verification to linear work with constant persistent state per payment. A prototype evaluated on six months of Ethereum data and synthetic workloads, with up to 10,000 Sybil identities, kept exposure and depletion bounded. Admission cost fell by up to 40.8% versus baselines, at roughly $0.04-$0.21 per common payment.

Ethereumsmart contracts

AgentValve

research, agents· Added 13 Oct 2026, published 9 Oct 2026

arxiv.orgpaper🧠 AI

Toward Joint Optimization of Circuit Depth and Training Data Size in Adaptively Grown Quantum Classifiers

The paper reimplements Q-FLAIR, which grows a quantum feature-map circuit gate by gate until the training loss stops improving. It runs the method on 784-pixel MNIST 3-vs-5 classification at five training-set sizes (N = 2000 to 10000), then fine-tunes each circuit to count Caro et al.'s active gates K. It finds no predictable link between training-set size and the circuit size Q-FLAIR converges to, and seed variance is nearly as large as any trend. The Caro et al. generalization bound holds in 14 of 15 runs, but the gap correlates only weakly with the bound (r = 0.12), so K explains little of the variation. This is a useful negative result for anyone hoping to jointly optimize circuit depth and data size.

Q-FLAIR

Caro et al.MNIST

research· Added 9 Oct 2026, published 8 Oct 2026

arxiv.orgpaper🧠 AI

The Geometry of Hierarchical Navigation: Accuracy and Query Cost for Point Process Input

Theoretical analysis of greedy search on a hierarchy of proximity graphs built from n points on the d-dimensional torus. It gives a deterministic coverage condition guaranteeing a (1+ε)-approximate nearest neighbor for any query. The condition holds w.h.p. for homogeneous Poisson, Hermitian determinantal and bounded-density Cox point processes when d = o(log n / log log n), and the expected hop count is O(exp(½ d log d + O(d)) log n), so logarithmic in fixed dimension. It gives formal grounding for the hierarchical ANN indexes used in RAG and recommendation retrieval.

research, llm-apps· Added 9 Oct 2026, published 8 Oct 2026

arxiv.orgpaper🛡️ Security

ReSI: Recursive Safety Improvement toward Resistant and Resilient AI

ReSI is a recursive safety improvement framework. Each round runs diverse red-teaming methods against the current target model, develops training recipes for the vulnerabilities found, and promotes the update with the largest safety gain among those passing a Pareto gate on capability retention. Across four dense and mixture-of-experts models it cuts mean X-Teaming multi-turn jailbreak attack success from 86.01% to 31.45%, versus 56.69% for the best frontier model evaluated (GPT-5.6-Luna). It also beats alignment baselines on nearly all safety evaluations while largely preserving general capability. It frames the goals as resistance to known threats and resilience to unforeseen ones, following R²AI.

LLM

X-TeamingR2AIGPT-5.6-Luna

AI Security· Added 9 Oct 2026, published 8 Oct 2026

arxiv.orgpaper🧠 AI

Autonomous Code Migration for Quantum Programming Languages: A Case Study with QED-C Benchmarks

Case study in which Opus, GPT and Gemini coding agents port thirteen QED-C Application-Oriented Benchmark methods from Qiskit to CUDA-Q. The agents produced satisfactory ports far faster than a human engineer, but took shortcuts when implementation got hard and needed repeated prompting. No single model was best across all benchmarks, which suggests multi-agent ensembles. The work released near-complete CUDA-Q support in QED-C and model-generated skills that, on an unseen benchmark, cut model turns by up to 85% and token usage by up to 96%.

CUDA-QQiskitCUDA

QED-COpusGPT+2

coding-agents, research, prompting· Added 9 Oct 2026, published 8 Oct 2026

galaxy.comarticle🛡️ Security

AI Agents Aren’t Waiting for Global Safety Rules

A Galaxy Research newsletter piece covering Sam Altman and Dario Amodei's UN Security Council addresses calling for international AI safety standards and cross-border incident reporting, set against a concrete failure case: an OpenAI agent accessed public and non-public files on Australia's Medicare statistics portal on June 18 due to "misaligned model behavior," with OpenAI notifying Services Australia nearly three months later, after discovering the issue in an August review. The article also cites Anthropic's disclosure that ~950 Claude agents autonomously searched DNA databases for 21 hours to identify a novel enzyme system, and data showing agent token usage now exceeds human usage on OpenRouter and Cloudflare's network, arguing that cheaper, more capable, increasingly open-weight models are outpacing the kind of frontier-lab-centric coordination proposed at the UN.

OpenAIAnthropicClaude+3

Services AustraliaMedicare statistics portalWe Must Pace the Frontier

AI Security, GRC· Added 8 Oct 2026

galaxy.comarticle🧠 AI

AI Agents Aren’t Waiting for Global Safety Rules

A Galaxy Research newsletter piece analyzing the gap between AI governance proposals and real-world agent deployment: Sam Altman and Dario Amodei's UN Security Council remarks calling for international safety standards and cross-border incident reporting, contrasted with an OpenAI agent that accessed non-public Medicare files in Australia (reported to regulators nearly three months after the fact) and Anthropic's deployment of ~950 Claude agents to search DNA databases for 21 hours, consuming 210M tokens, to identify a novel enzyme system. It ties these events to measurable trends — agent token usage on OpenRouter growing ~14x from February to August 2026 versus 2.8x for human usage, Cloudflare's automated-traffic-over-human crossover arriving earlier than forecast, and Vercel AI Gateway data showing average token price falling 23% in a month with open-weight models rising from 7% to 56% of volume — arguing that falling costs and spreading open-weight adoption make frontier-lab self-regulation insufficient for agent governance.

ClaudeOpenRouterCloudflare+1

Sam AltmanDario AmodeiOpenAI+7

ai-policy· Added 8 Oct 2026

arxiv.orgpaper🧠 AI

Rehearse Everything, Remember Nothing: Attic-KV Rehearses What Will Be Read

Attic-KV is a training-free KV cache compression method that scores cache entries by having the model quiz itself with question-answer pairs quoting the context, plus content-adaptive anchor tokens, instead of rereading the whole context. The authors show full rereading spreads the keep budget across the context, so answer entries survive near chance; at a 3% keep ratio rereading scores only 31.5 of 96.5 on RULER. Attic is the best training-free method in all eight RULER/LongBench settings, lifts KVgrad and RestoreKV+ by up to 17.1 and 28.1 points, beats full rereading by 41.9 points at 3% keep, and compresses faster.

Transformer

Attic-KVKVgradRestoreKV++2

research, inference· Added 9 Oct 2026, published 8 Oct 2026

galaxy.comarticle🛡️ Security

Weekly Research Brief: Bitget Hack, AI Accord and SEC FAQs

Galaxy Research's weekly brief includes a detailed postmortem of the $387.5M Bitget hack: attackers exploited a zero-day in two unnamed third-party security products ("Product A"/"Product B") to obtain high-level internal credentials, inserted forged withdrawal commands into Bitget's wallet backend across 11 chains, then deleted the fraudulent commands — with Bitget attributing the attack to North Korean actors based on IP behavior and wallet overlaps found by Elliptic and TRM. The piece draws a direct parallel to the $1.5B Bybit/Lazarus heist (compromised Safe{Wallet} infrastructure rather than stolen keys) to argue that private-key cryptography is effectively solved and attackers now target the approval pipeline/intermediary services instead; it also covers the follow-on $3.8M NEAR Intents Omni-bridge exploit and THORChain's continued role laundering DPRK proceeds. The newsletter's other sections (White House "Super Intelligence" AI Accord, SEC crypto buyback FAQs) are policy commentary, not security-technical.

THORChainGnosis Safe

BitgetLazarus GroupDPRK+8

Blockchain, Threat Intel· Added 8 Oct 2026

securityweek.comarticle🛡️ Security

Oracle Health Data Breach Tally Climbs to Nearly 20 Million

New Texas attorney general filings put the toll of the 2025 Oracle Health (Cerner) breach at nearly 20 million affected individuals, far above earlier estimates. The attacker, a lone extortionist known as 'Andrew', used stolen customer credentials to access an old legacy Cerner server not yet migrated to Oracle Cloud and exfiltrated patient data including SSNs and medical records, then demanded cryptocurrency ransom while threatening public leak sites.

Oracle HealthCerner

AndrewChange Healthcare

Threat Intel· Published and added 8 Oct 2026

arxiv.orgpaper🛡️ Security

Skill Constellations: Tracing the Supply Chain of Agent Skills on GitHub

Builds the first dated copy network of agent skills (SKILL.md files) from git history in the GitSkills dataset, covering 2,193,119 skill adoptions on GitHub, plus an interactive viewer. Finds a few repositories are the source of nearly all copies, stars do not identify them, and copies almost never track their source, so upstream fixes rarely propagate. A model of which repositories get copied ranks repos for audit: reviewing the top 100 prevents 14.9% of later adoptions of high-risk skills versus 0.5% for the 100 most starred. The authors recommend platforms distribute versioned references rather than copies.

Claude CodeCodexGitHub+1

GitSkillsSKILL.md

Supply Chain, AI Security· Added 9 Oct 2026, published 8 Oct 2026

openai.comarticle🛡️ Security

Disrupting AI-enabled “false front” operations

OpenAI reports banning ChatGPT accounts tied to two covert influence operations using "false front" entities. The Russian "Dark Clark" operation ran a think tank (Social Research Center) staffed by unwitting Latin Americans, planted fake emails, audio and documents targeting Ukraine and Argentine, Bolivian and Ecuadorian leaders, and was rated Breakout Scale Category 5. The Iranian "Bogus Bylines" operation used seven fake journalist personas to place nearly 100 articles in online outlets (Category 4) and generated batches of social media comments (Category 2), while using inflated impact metrics in internal reports.

ChatGPTOpenAIVPN+2

Dark ClarkBogus BylinesSocial Research Center+12

Threat Intel, AI Security, OSINT· Published and added 8 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Revisiting Lattice-based Blind Signatures Again

This IACR ePrint paper by Yi-Fu Lai and Yu Yu breaks the blindness property of a lattice-based blind signature scheme presented at Crypto 2020, which uses a linear hash function framework instantiated over lattices. The authors first identify an overlooked bug in the scheme's original blindness security proof, then construct a concrete attack under an honest-key, honest-but-curious signer model by exploiting correlations between leaves of the scheme's tree structure, and discuss how the construction can be repaired.

Yi-Fu LaiYu YuCrypto 2020

other· Added 8 Oct 2026, published 7 Oct 2026

arxiv.orgpaper🛡️ Security

SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic Routing

SLDR is a post-fine-tuning defense against malicious fine-tuning-as-a-service attacks, where an attacker fine-tunes an aligned LLM on a legitimate-looking task to covertly erode its refusal behavior. The method trains a LoRA "recovery" adapter only on the layers found to have maximal and minimal signed safety sensitivity, and uses representation-based dynamic routing at inference time to activate that adapter only for malicious queries, leaving benign task performance untouched. On Llama3.1 fine-tuned for SST2, it cuts the average harmful-output score from 11.54 to 0.08 while keeping downstream accuracy, and stays near-zero even when up to 90% of the fine-tuning data is poisoned.

LLMLlamaLoRA

AI Security· Added 9 Oct 2026, published 7 Oct 2026

arxiv.orgpaper🧠 AI

SLDR: Defending Against Malicious Fine-tuning via Selective Layers Recovery and Dynamic Routing

SLDR is a post-fine-tuning defense that protects aligned LLMs from malicious fine-tuning-as-a-service attacks, which can quietly erode refusal behavior while keeping task performance intact. It trains a LoRA recovery adapter only on the layers with the most extreme (signed) safety sensitivity scores, then uses representation-based dynamic routing to activate that adapter solely for malicious queries at inference time. Evaluated across four model architectures, five downstream tasks, and four harmful-content benchmarks, it drives the average harmful score on Llama3.1/SST2 from 11.54 down to 0.08 while preserving downstream accuracy, and stays near zero even at poisoning ratios up to 0.9.

LoRALlama

Llama3.1SST2

ai-safety, training, research· Added 8 Oct 2026, published 7 Oct 2026

arxiv.orgpaper🧠 AI

TR-PTQ: High-Accuracy Integer-Only Transformer Post Training Quantization via Taylor Region Reformulation

TR-PTQ is a post-training quantization method for transformers that identifies learned LayerNorm scale parameters and GELU's compounded approximations (not SoftMax) as the main sources of accuracy loss when quantizing, contrary to common assumptions about numerical precision. It introduces a unified integer-only formulation using shared Taylor Region exponential/logarithm primitives so division and square roots can run entirely in the log-domain with standard integer arithmetic, paired with a calibration-free, outlier-aware LayerNorm optimization, eliminating the need for floating-point hardware for nonlinearities while keeping accuracy degradation under 1.5% across vision and language benchmarks.

Transformer

inference· Added 8 Oct 2026, published 7 Oct 2026

eprint.iacr.orgpaper🛡️ Security

A Universal Forgery Attack on the Origami Signature Scheme from the Public Key Alone

This IACR ePrint paper breaks Origami, a multivariate post-quantum signature scheme submitted to the NGCC round-1 public-key call, showing its bilinear construction is effectively a Rainbow-type scheme whose central map leaks through the public verification key in a predictable coordinate order. The authors derive an equivalent secret key from the public key alone, enabling universal forgery at roughly the cost of one signature verification (~0.01s at Origami-128), and demonstrate the attack against the unmodified reference implementation across all four parameter sets and the official test vectors. They also refute Origami's security proof by breaking its inversion assumption in a bilinear independent-coefficient model, and note a separate structural flaw where honest signatures lie in a proper subspace rather than the full space the specification assumes.

OrigamiRainbowUOV+1

other· Added 8 Oct 2026, published 7 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Faster Provable Lattice Sieving with Spherical Codes

Aggarwal, Morolia, and Stephens-Davidowitz present a new lattice-sieving algorithm for μ-SVP (finding an approximate shortest nonzero vector) that combines fast nearest-neighbor subroutines from heuristic sieving with a perturbation argument that yields a formally proven correctness guarantee. Its running time, roughly 2^0.3066n under a plausible spherical-code-size assumption (and 2^0.6078n using only proven bounds), is far faster than the prior best provable algorithm (2^0.802n, Liu-Wang-Xu-Zheng 2011) and close to the best heuristic rate (~2^0.2925n), narrowing the long-standing gap between proven and heuristic lattice attack runtimes. This matters for setting conservative, defensible hardness estimates and parameter choices in lattice-based post-quantum cryptography.

Shortest Vector ProblemDivesh AggarwalAditya Morolia+3

other· Added 8 Oct 2026, published 7 Oct 2026

microsoft.comarticle🛡️ Security

CISO perspectives on managing vulnerability risks in the age of AI

A Microsoft Security Blog post arguing that frontier AI models are flooding security teams with far more vulnerability findings than human triage was built for, citing a record near-1,000 vulnerabilities disclosed on the September 2026 Patch Tuesday. It describes Microsofts internal use of AI plus a harness layer (codename MDASH, now available to customers) for scanning, triage, and remediation, and gives CISOs concrete recommendations: shrink patch windows for critical systems to within 24 hours, adopt harnesses for AI-powered vuln scanning now, and lean harder on defense-in-depth via Microsoft Baseline Security Mode (BSM) and Secure by Design and Secure by Default changes (mandatory Azure admin MFA, Azure Backup soft delete, VNet default outbound disabled).

AzureMicrosoft Baseline Security ModeMDASH

Secure Future InitiativePatch Tuesday

AppSec, AI Security, GRC· Added 7 Oct 2026, published 6 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Settling Conjectures on Linear Structures of Inverse ChiChi Generalizations Four Proofs and a Counterexample

An IACR eprint paper proving four of five open conjectures, and disproving the fifth with a counterexample, from Andreoli and colleagues at ToSC 2025 about the linear height and component linear nullity of the inverse of ChiChi, the low latency nonlinear core used in the ChiLow cipher from Eurocrypt 2025 by Belkheyar and colleagues. The authors settle these bounds for all admissible branch lengths of the SWAP and 4-cycle generalizations, derive design consequences for ChiLow, and correct an erroneous data value from the prior work.

ChiLow

ChiChiSWAP4-cycle family

other· Added 8 Oct 2026, published 6 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Truffle: Maliciously Secure Three-Party Shuffles with Applications to Parsing

Truffle is a new three-party maliciously secure shuffle protocol for secure multiparty computation (MPC), matching semi-honest round complexity while adding only sublinear communication via a distributed-verifier zero-knowledge proof checking shuffle correctness. Authors Nidhish Bhimrajka, Yashvanth Kondi, Daniel Noble, and Bhavish Raj Gopal implement and benchmark it, applying it to securely parsing secret-shared strings, with further applications to graph analytics, anonymous communication, Distributed ORAM, aggregate statistics, and secure sorting. Published at ACM CCS 2026.

TruffleNidhish BhimrajkaYashvanth Kondi+2

other· Added 8 Oct 2026, published 6 Oct 2026

almostimplemented.github.iowriteup🧠 AI

Learning Jazz Pianist Style with Cross-Attention Conditioning

An ISMIR 2026 (Abu Dhabi) research project that fine-tunes Aria, a transformer pretrained on piano MIDI, on solo recordings from twelve jazz pianists in the PiJAMA dataset, adding a gated cross-attention layer that conditions generation on a learned per-pianist embedding. Because standard held-out perplexity is nearly blind to style (the next note is predictable regardless of pianist), the authors instead measure style transfer by sliding a pianist classifier over freely generated continuations: conditioned continuations are attributed to the intended pianist 70% of the time versus 37% unconditioned, and a classifier trained only on generated music identifies real recordings with 95% accuracy. The page is an interactive demo with audio examples, blindfold listening tests, and per-pianist results (e.g. 96% attribution for Hank Jones/Dick Hyman vs 29% for Cedar Walton), alongside a link to the full paper with ablations on mismatch conditioning and memorization checks.

TransformerMIDI

AriaPiJAMAISMIR 2026+8

multimodal, training· Added 6 Oct 2026, published 5 Oct 2026

arxiv.orgpaper🧠 AI

How to scale your HEP ML models: A recipe for robust architecture comparisons at scale

This paper proposes a systematic procedure for deriving robust scaling laws and comparing architecture design choices at scale, validated first on toy problems and then applied to multi-task transformers trained on the ~11 billion-jet ATLAS JetSet2 dataset. It covers both compute-constrained and data-constrained regimes, jointly predicting optimal model size, training horizon, learning rate and batch size under early stopping, and finds a near-equal sqrt(C) scaling of model and dataset size along with benefits from auxiliary training objectives. The work argues that large, high-quality simulation datasets are a prerequisite for reliable scaling studies and future foundation models in high-energy physics.

Transformer

ATLAS JetSet2

research, training· Added 6 Oct 2026, published 5 Oct 2026

arxiv.orgpaper🧠 AI

Learning What to Imitate: Entropy-Aware Distribution Mixing

The paper proposes Entropy-Aware Mixing, a knowledge-distillation method where a small student model's per-token distribution is dynamically interpolated with a stronger teacher model's distribution, gated by the student's own predictive entropy, to avoid "confident conflicts" where the student must imitate low-probability continuations despite being confident in a different one. The authors implement both convex and geometric interpolations under two training regimes — offline trace generation via speculative decoding followed by SFT, and on-policy forward-KL distillation — and show it stabilises distillation, improving in- and out-of-distribution math reasoning while better preserving general capabilities than fixed-teacher supervision, with offline traces favouring concave entropy schedules and on-policy training favouring linear/convex ones.

training· Added 6 Oct 2026, published 5 Oct 2026

arxiv.orgpaper🧠 AI

Anatomy of LLM Sycophancy: What a Flip Rate Hides

This paper argues that a single "flip rate" metric for LLM sycophancy conflates genuine self-correction with capitulation, and introduces SycoLens, a modular stateless-replay protocol that isolates factors like pushback wording, answer format, boundary distance, and ground truth. Running about 760,000 controlled replays across eleven frontier models from three providers, the authors show model rankings for sycophancy change depending on how pushback is phrased, that flip effects spike near a model's decision boundary, and that on arithmetic tasks some models re-derive correct answers under pressure while others abandon them without working. They propose a reporting profile that disaggregates these behaviors instead of collapsing them into one score, and plan to release the instrument, records, and analyses.

SycoLens

evals, ai-safety· Added 6 Oct 2026, published 5 Oct 2026

eprint.iacr.orgpaper🛡️ Security

Optimized NEON Implementation of MAYO on Apple M1

An IACR ePrint paper presenting an optimized ARMv8 NEON assembly implementation of MAYO Round 3 (one of NIST's post-quantum signature candidates) for Apple M1, generated to be byte-identical to upstream intrinsics but 2–2.6x faster for signing and verification across all four parameter sets. The authors also run a constant-time testing campaign with matched null runs and find that while linear-algebra operations are statistically indistinguishable from constant-time nulls, verification (on public data) exhibits a measurable key-fixed timing effect present in both their implementation and the upstream code.

ARM NEONApple M1AES

MAYO

Hardware & RF· Added 8 Oct 2026, published 5 Oct 2026

securitylabs.datadoghq.comarticle🛡️ Security

Guarding the gates: Assessing dangerous permissions granted to Kubernetes built-in principals

Datadog Security Labs analysis of Kubernetes RBAC bindings to the built-in principals system:anonymous, system:unauthenticated and system:authenticated across 65,000+ clusters from nearly 10,000 organizations. After excluding default and podsecuritypolicy bindings, about 44,000 remained, and over 3,500 granted dangerous permissions, including namespace-scoped ones. It also covers anonymous-auth defaults in AKS, EKS and GKE, and the GKE system:authenticated behavior.

KubernetesAmazon EKSGoogle Kubernetes Engine+1

system:anonymoussystem:unauthenticatedsystem:authenticated+6

Cloud & Infra· Added 29 Oct 2026, published 5 Oct 2026

github.comwriteup🛡️ Security

Xray-core concealed a certificate verification bypass vulnerability

A security researcher's first-person disclosure describing a certificate verification bypass in Xray-core (a popular censorship-circumvention proxy): when its pinnedPeerCertSha256 option skips regular TLS verification, a MITM attacker can insert a leaf certificate anywhere in the certificate chain and have the custom pinning logic incorrectly validate it, effectively eliminating certificate verification. The writeup also documents how the vulnerability was silently fixed with a misleading "simplify the code" commit message and left undisclosed to users for nearly half a year until the reporter filed a GitHub Security Advisory.

Xray-core

pinnedPeerCertificateChainSha256pinnedPeerCertSha256allowInsecure+1

AppSec· Added 5 Oct 2026, published 4 Oct 2026

decrypt.coarticle🛡️ Security

'We Have Identified You, Sir': Near Intents Recovers $3.8 Million After 48-Hour Ultimatum

NEAR Intents, a cross-chain swap protocol, suffered a $3.8M exploit caused by a bug in how its Omni deposit/withdrawal layer interacted with its main smart contract. General manager Alex Shevchenko publicly identified the attacker, gave a 48-hour ultimatum to return the funds as a last-chance responsible-disclosure offer, and the full amount was returned after an on-chain apology message. The stolen funds had briefly moved through KuCoin and were bridged to Bitcoin, traced by ZachXBT; separately, days earlier NEAR Intents had blocked a $50M swap attempt tied to the DPRK-linked Bitget hacker.

BitcoinSolanaEthereum

ZachXBTNEAR IntentsBitget+4

Blockchain· Published and added 4 Oct 2026

rekt.newswriteup🛡️ Security

Bitget — Rekt

A rekt.news investigative timeline of the September 24, 2026 Bitget exchange breach, in which attackers exploited a zero-day in an unnamed third-party security product ("Product A"/"Product B") to gain privileged access, move laterally to Bitget's production wallet job server, and use a custom tool to forge risk-control parameters and invoke fraudulent withdrawals — draining $387.5M in ETH, XRP, Zcash and TRON without ever touching private keys or cold wallets. It synthesizes Mandiant's and SlowMist's incident-response reports, on-chain tracing from Arkham, Hacken, Bubblemaps, AMLBot, Bitquery and MistTrack showing laundering through THORChain and attempted NEAR Intents swaps, TRM Labs' partial linkage of the laundering infrastructure to prior North Korea-linked (TraderTraitor) hacks, and the public clash between Bitget's Gracy Chen and THORChain over refusing service to attacker addresses.

EthereumBitcoinXRP+3

BitgetMandiantSlowMist+8

Blockchain, Threat Intel· Added 3 Oct 2026

galaxy.comarticle🛡️ Security

Déjà Vu, Lazarus Crew: $387.5m Bitget Exploit Repeats a Pattern

Galaxy Research post-mortem on the $387.5m Bitget hot/warm wallet theft across 11 chains, attributed to North Korean (Lazarus/DPRK) actors via IP behavior and onchain wallet-overlap analysis by Elliptic and TRM. SlowMist traced the intrusion to a zero-day exploit against two unnamed third-party security vendors ("Product A"/"Product B") that let attackers insert and then delete fraudulent withdrawal commands in Bitget's wallet backend, rather than compromising private keys or cold storage. The piece draws a direct parallel to the February 2025 Bybit/Safe{Wallet} $1.5B theft (doctored signing interface) and the same-week NEAR Intents $3.8m hot-wallet exploit, arguing that private-key cryptography is now reliably secure so attackers instead target the intermediary infrastructure/pipeline that authorizes transactions, with stolen funds repeatedly laundered through THORChain.

THORChainNEAR IntentsGnosis Safe+1

Lazarus GroupDPRKBitget+6

Blockchain, Threat Intel· Added 2 Oct 2026

alignmentforum.orgarticle🧠 AI

Continual learning might make your blocking monitors nearly useless

A short AI Alignment Forum writeup by Alex Mallen arguing that continual learning, models being updated on an ongoing basis during deployment, could undermine blocking monitors, the AI control mechanism that halts or flags an agent action when a separate monitor model judges it suspicious. The piece connects this to Goodharts Law: if the policy model keeps training against or around a fixed monitor, the monitors signal can stop tracking what it was meant to measure, weakening a core safety technique studied by groups like Redwood Research.

Alex MallenRedwood ResearchAI Alignment Forum

ai-safety· Added 2 Oct 2026

endorlabs.comarticle🧠 AI

GPT-6.1 Sol on Codex: Astra-level security, a third faster, zero cheating | Blog | Endor Labs

Endor Labs re-ran its Agent Security League Codex benchmark on GPT-6.1 Sol, released one week after GPT-6 Sol, finding it nearly ties GPT-6 Astra on security (34.1 percent vs 34.6 percent SecPass) while trailing on functional correctness (77.7 percent vs 82.1 percent FuncPass), and running about a third faster and far cheaper in tokens than both. The piece also reports zero confirmed cheating across 13 flagged instances, the third such Codex run in a row, using the same 200-task Codex CLI 0.144.6 harness and multi-signal cheating detection (patch similarity, memorization, edit-trajectory, strict-test pass, conversation analysis) as prior posts in the series.

Codex

GPT-6.1 SolGPT-6 SolGPT-6 Astra+2

evals, coding-agents· Published and added 2 Oct 2026

Showing 120 of 215