LoRA Without Regret
Thinking Machines Lab · Thinking Machines
The paper’s “low-regret regime” is conditional: LoRA can match full fine-tuning for post-training when adapters cover important layers and have enough capacity.
Open library
The library
The essential reading of generative AI and agents, open in a clean reading view, with Meow and Standard summaries when available. No account needed to read.
560 papers · 558 with a Meow summary · 333 posts
Key publications from Anthropic, OpenAI, Google DeepMind, Meta, DeepSeek, Qwen, Z.ai, Kimi and other labs, plus the most influential essays and blogs.
On-Policy Distillation
Thinking Machines Lab · Thinking Machines
On-policy distillation has the student generate its own trajectories while a strong teacher gives token-level feedback—combining RL’s student-specific training with distillation’s denser signal.
Thinking about High-Quality Human Data
Lilian Weng
A cleaner label set is not always a better one: quality work must remove annotation errors without flattening legitimate disagreement.
RLHF: Reinforcement Learning from Human Feedback
Chip Huyen
RLHF uses human rankings to steer a language model toward preferred answers. InstructGPT’s SFT+RLHF model outperformed SFT alone and was preferred overall, but hallucinated more.
New LLM Pre-training and Post-training Paradigms
Sebastian Raschka, PhD · Sebastian Raschka
Across these reports, the clearest trend is deliberate data quality and staged training—not one winning recipe—with teams extending context and customizing alignment.
RLHF Book: Reinforcement Learning from Human Feedback and LLM Post-Training
Nathan Lambert
This page maps web edits against changes carried into the October 2026 first reprint, helping readers tell whether the print and web versions match.