Linfeng Song

Linfeng Song on Hugging Face Daily Papers: 13 papers, 2 in the top 3 of their day, 305 upvotes.

  1. Towards Solving More Challenging IMO Problems via Decoupled Reasoning and Proving 15 upvotes, #10 of 2025-07-10
  2. DeepTheorem: Advancing LLM Reasoning for Theorem Proving Through Natural Language and Reinforcement Learning 15 upvotes, #24 of 2025-05-30
  3. MPS-Prover: Advancing Stepwise Theorem Proving by Multi-Perspective Search and Data Curation 7 upvotes, #9 of 2025-05-19
  4. DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning 11 upvotes, #17 of 2025-04-16
  5. Expanding RL with Verifiable Rewards Across Diverse Domains 17 upvotes, #11 of 2025-04-01
  6. Thoughts Are All Over the Place: On the Underthinking of o1-Like LLMs 51 upvotes, #2 of 2025-01-31
  7. HUNYUANPROVER: A Scalable Data Synthesis Framework and Guided Tree Search for Automated Theorem Proving 10 upvotes, #4 of 2025-01-02
  8. Do NOT Think That Much for 2+3=? On the Overthinking of o1-Like LLMs 27 upvotes, #5 of 2024-12-31
  9. Towards Self-Improvement of LLMs via MCTS: Leveraging Stepwise Knowledge with Curriculum Preference Learning 9 upvotes, #15 of 2024-10-11
  10. LiteSearch: Efficacious Tree Search for LLM 34 upvotes, #4 of 2024-07-02
  11. Iterative Nash Policy Optimization: Aligning LLMs with General Preferences via No-Regret Learning 6 upvotes, #8 of 2024-07-01
  12. Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing 44 upvotes, #1 of 2024-04-19
  13. Stabilizing RLHF through Advantage Model and Selective Rehearsal 10 upvotes, #6 of 2023-09-20

Data: hysts-bot-data/daily-papers-stats and the Daily Papers API. Open data: tardellirs/paper-pulse-data. Sister project: Model Pulse, the download history of every model on the Hub.