Thomas Wolf PRO

thomwolf

·

https://thomwolf.io

AI & ML interests

NLP and open-source :-)

Recent Activity

liked a Space about 2 hours ago

lvwerra/agent-manager-template

new activity about 2 hours ago

rl-llm-wiki/knowledge-base:source: arxiv:2507.18071 — GSPO (sequence-level IS ratio + clipping; the Qwen3 RL loss)

new activity about 2 hours ago

rl-llm-wiki/knowledge-base:source: arxiv:2207.14502 — LMs Can Teach Themselves to Program Better (verifier-filtered self-improvement)

View all activity

Organizations

liked a Space about 2 hours ago

Agent Manager

Private cloud manager for AI coding CLI sessions

New activity in rl-llm-wiki/knowledge-base about 2 hours ago

source: arxiv:2507.18071 — GSPO (sequence-level IS ratio + clipping; the Qwen3 RL loss)

#373 opened about 2 hours ago by

source: arxiv:2207.14502 — LMs Can Teach Themselves to Program Better (verifier-filtered self-improvement)

#369 opened about 8 hours ago by

topic: algorithms/grpo-and-group-relative — developing → comprehensive

#365 opened about 10 hours ago by

fix: self-distillation-and-rich-feedback-rl — precise Qwen2.5 scale-floor wording (post-#363 review note)

#372 opened about 2 hours ago by

source: arxiv:2506.13585 — MiniMax-M1 (CISPO: clipped IS-weight policy optimization + RL stability recipes)

#371 opened about 2 hours ago by

source: arxiv:2510.13786 — The Art of Scaling RL Compute for LLMs (ScaleRL; sigmoid compute-scaling framework, CISPO adoption)

#370 opened about 2 hours ago by

updated a bucket about 2 hours ago

rl-llm-wiki/rl-milo-the-cat

updated a Space about 3 hours ago

RL-for-LLMs Dashboard

Agents collaboratively build an expert-level wiki

updated a Space about 15 hours ago

RL-Town

The RL-for-LLMs collaboration, replayed as a living town

upvoted a changelog about 16 hours ago

Hugging Face Changelog

Share your feedback with us

8 days ago

• 110

liked a model about 16 hours ago

zai-org/GLM-5.2

Text Generation • 753B • Updated 2 days ago • 209k • • 3.37k

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

source: arxiv:2601.20802 — SDPO (RL via Self-Distillation) + new node algorithms/self-distillation-and-rich-feedback-rl

#363 opened about 16 hours ago by

published a bucket about 16 hours ago

rl-llm-wiki/rl-milo-the-cat

liked a Space about 17 hours ago

RL-Town

The RL-for-LLMs collaboration, replayed as a living town

published a Space about 17 hours ago

RL-Town

The RL-for-LLMs collaboration, replayed as a living town

liked a model 1 day ago

katanemo/Plano-Orchestrator-4B

Text Generation • 4B • Updated Apr 2 • 41 • • 13

liked a model 3 days ago

Qwen/Qwen3-TTS-12Hz-1.7B-CustomVoice

Text-to-Speech • 2B • Updated Jan 29 • 2M • 1.67k

liked a Space 3 days ago

HF Realtime Voice

Voice chat over WebSocket against a HF speech-to-speech

upvoted an article 3 days ago

Article

Hugging Face and Cerebras bring Gemma 4 to real-time voice AI

+2

A-Mahla, andito, lvwerra, vyassaurabh

•

4 days ago

• 53