StepFun-Formalizer: Unlocking the Autoformalization Potential of LLMs through Knowledge-Reasoning Fusion (2025 · 7 citations)
Autoformalization means converting natural-language math problems into formal statements, here in Lean 4. Current LLM approaches are still inaccurate. General-purpose models often misuse Lean details, such as calling functions that do not exist. Specialized translators trained on narrow annotated data tend to misread…
Rhombus: Incentivizing Coordination in Parallel Thinking through Reinforcement Learning (2026)
Parallel reasoning lets an LLM explore several solution attempts at once and then merge them. In practice, the explorers and the merger are usually built separately. The explorers do not know what the aggregator needs, so useful intermediate evidence gets lost at the hand-off. Rhombus treats this hand-off as a…

All works

3 works of theirs in our library, the most cited first; 2 with a page here, the others link their original until we summarise them.

StepFun-Formalizer: Unlocking the Autoformalization Potential of LLMs through Knowledge-Reasoning FusionarXiv:2508.04440Yutong Wu, Di Huang, Ruosi Wan, et al.2025 · 7 citations
Rhombus: Incentivizing Coordination in Parallel Thinking through Reinforcement LearningFindings of the Association for Computational LinguisticsZiyuan Nan, Qi Yi, Di Huang, et al.2026
Revisiting Entropy Regularization: Adaptive Coefficient Unlocks Its Potential for LLM Reinforcement Learning ↗Findings of the Association for Computational Linguistics: ACL 2026 · no summary yetXiaoyun Zhang et al.2026

Co-authors here

9 co-authors with a page, the most papers together first.

Where their works here were published, the most works first, with the editions of those that have a page.

Papers of our library cite their works 7 times (of the 149,439 papers whose bibliographies we read).

Updated 1 November 2026