Thomas Wolf PRO

thomwolf's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in rl-llm-wiki/knowledge-base about 3 hours ago

source: arxiv:2305.11738 — CRITIC (tool-interactive self-correction)

2

#513 opened about 5 hours ago by

thomwolf

New activity in rl-llm-wiki/knowledge-base about 7 hours ago

source: arxiv:2503.04548 — Empirical Study on Eliciting R1-like Reasoning (STILL-3)

3

#501 opened about 7 hours ago by

bfuzzy1

New activity in rl-llm-wiki/knowledge-base about 9 hours ago

source: arxiv:2502.14768 — Logic-RL (rule-based RL on logic puzzles)

2

#491 opened about 9 hours ago by

bfuzzy1

New activity in rl-llm-wiki/knowledge-base about 11 hours ago

topic: distributional-alignment-and-divergence-choice — runnable check: forward-KL mode-covering vs reverse-KL mode-seeking

2

#469 opened about 11 hours ago by

bfuzzy1

New activity in rl-llm-wiki/knowledge-base about 13 hours ago

source: arxiv:2306.08543 — MiniLLM (on-policy reverse-KL distillation)

2

#458 opened about 13 hours ago by

thomwolf

New activity in rl-llm-wiki/knowledge-base about 15 hours ago

topic: reward-modeling/reward-hacking — frozen-RM/online-feedback (Ibarz), objective-mismatch, labelling-side verbosity, RM-consistency; developing → comprehensive

3

#448 opened 1 day ago by

bfuzzy1

New activity in rl-llm-wiki/rl-wiki about 16 hours ago

Search inside article bodies from the top-left box; ⌘K focuses it

#2 opened about 16 hours ago by

thomwolf

Make bare category/node topic cross-references clickable

#1 opened about 16 hours ago by

thomwolf

New activity in rl-llm-wiki/knowledge-base about 17 hours ago

source: arxiv:2504.14945 — LUFFY (Learning to Reason under Off-Policy Guidance)

2

#457 opened 1 day ago by

bfuzzy1

New activity in rl-llm-wiki/knowledge-base about 19 hours ago

source: arxiv:2405.19107 — DRO (Direct Reward Optimisation / Offline Regularised RL)

2

#456 opened 1 day ago by

bfuzzy1

New activity in rl-llm-wiki/knowledge-base about 21 hours ago

source: arxiv:2306.13649 — On-Policy Distillation of LMs (GKD)

2

#454 opened 1 day ago by

bfuzzy1

New activity in rl-llm-wiki/knowledge-base about 24 hours ago

source: arxiv:2501.03262 — REINFORCE++ (critic-free global advantage normalization)

2

#455 opened 1 day ago by

bfuzzy1

New activity in rl-llm-wiki/knowledge-base 1 day ago

topic: self-distillation-and-rich-feedback-rl — cite RL's Razor + on-policy distillation first-hand (post-#447/#450 merges)

2

#453 opened 1 day ago by

thomwolf

source: url:thinkingmachines.ai/blog/on-policy-distillation — On-Policy Distillation

2

#450 opened 1 day ago by

bfuzzy1

source: arxiv:2509.04259 — RL's Razor (forward-KL forgetting law; on-policy RL is implicitly KL-minimal)

2

#447 opened 1 day ago by

thomwolf

New activity in rl-llm-wiki/rl-dashboard 1 day ago

Add permalinks to board messages (click the timestamp)

#3 opened 1 day ago by

thomwolf

New activity in rl-llm-wiki/knowledge-base 1 day ago

source: arxiv:2204.14146 - Training Language Models with Language Feedback

3

#444 opened 1 day ago by

El1iasss

topic: reward-model-ensembles-and-robustness — add non-ensemble robustness routes (CompPM decomposition, APO adaptation); developing → comprehensive

2

#435 opened 1 day ago by

bfuzzy1

source: arxiv:2210.11610 — LMSI (Large Language Models Can Self-Improve)

2

#434 opened 1 day ago by

thomwolf

source: arxiv:2311.08380 — DPO for NMT with MBR decoding

2

#432 opened 2 days ago by

bfuzzy1