🎯 AI: techniques

Attack and defence techniques and machine learning methods that a second item could name too, with an ATT&CK, ATLAS or CWE id when one is stated. A procedure only one article describes is item text. Each one opens the items that name it, and the list is ranked by how many do.

This workspace's kept items name 295 of them. This is page 1 of 3, the most named first.

• Speculative decoding23 items
• Distillation17 items
• Mixture of experts17 items
• Retrieval-augmented generation15 items
• Chain-of-thought monitoring14 items
• Prompt caching13 items
• Quantization10 items
• LoRA9 items
• Computer use8 items
• Group Relative Policy Optimization8 items
• GRPO8 items
• Post-training quantization8 items
• Recursive self-improvement8 items
• Decision models7 items
• Quantization-aware training7 items
• RLVR7 items
• Supervised fine-tuning7 items
• Vibe coding7 items
• Context compaction6 items
• Loop engineering6 items
• LoRA fine-tuning6 items
• Progressive disclosure6 items
• Reinforcement learning6 items
• Reinforcement learning with verifiable rewards6 items
• Abliteration5 items
• Agent skills5 items
• Expert Parallelism5 items
• Looped Transformers5 items
• Matryoshka Representation Learning5 items
• Multi Round-Trip Requests5 items
• NVFP4 quantization5 items
• Preserved thinking5 items
• RAG5 items
• Tensor parallelism5 items
• Adam4 items
• Continuous Batching4 items
• Debate4 items
• Direct Preference Optimization4 items
• Distribution Matching Distillation4 items
• Mid-training4 items
• Multi-teacher on-policy distillation4 items
• Multi-Token Prediction4 items
• Pipeline parallelism4 items
• Plan Mode4 items
• RLCD4 items
• RLHF4 items
• Sandboxing4 items
• Skills4 items
• Attention Residuals3 items
• Calibration3 items
• Client ID Metadata Documents3 items
• Context engineering3 items
• Derived bindings3 items
• Derived data3 items
• Disaggregated prefill/decode serving3 items
• Distillation attack3 items
• DPO3 items
• Effort levels3 items
• fine-tuning3 items
• GEPA3 items
• Grouped Query Attention3 items
• Groupwise agentic grading3 items
• Kimi Delta Attention3 items
• Megakernels3 items
• Model routing3 items
• Multi-harness RL3 items
• On-policy distillation3 items
• PagedAttention3 items
• pass@13 items
• Persistent memory3 items
• Prefill-Decode disaggregation3 items
• Prefix caching3 items
• Red-teaming3 items
• Safe Completions3 items
• Self-Forcing3 items
• Sparse Mixture-of-Experts3 items
• SWA Bounded Replay3 items
• Tool calling3 items
• Activation probes2 items
• Activation steering2 items
• Adjustable effort setting2 items
• Adversarial distillation2 items
• Adversarial review2 items
• AFlow2 items
• Agent loop2 items
• Agentic Context Engineering2 items
• Agentic RL2 items
• AGENTS.md2 items
• Airgapping2 items
• Aligned RL2 items
• Alignment2 items
• Anchor as identity, URN as alias2 items
• APEX2 items
• Approval binding2 items
• Asynchronous CPU scheduling2 items
• Asynchronous KV loading via GDR/RDMA2 items
• Attention Interface2 items
• Automated behavioral audit2 items
• Automated red teaming and adversarial training2 items
• Base-branch policy reading2 items