研究人员正在探索先进的代理推理框架，以增强大型语言模型（LLM）在形式化验证和定理证明等复杂任务中的能力。诸如可验证过程奖励（VPR）之类的新方法旨在通过利用中间决策的客观检查来提供更密集、逐轮的监督，从而改进长时信用分配。还正在开发代理引导的树搜索和统计可证性理论，以优化证明生成并理解代理定理证明器中不同组件的有效性。这些进展在从数学推理到程序验证的领域都显示出希望，尽管在处理非结构化环境和开发更鲁棒的评估方法方面仍存在挑战。
AI
arXiv:2605.10325v2 Announce Type: replace Abstract: Reinforcement learning from verifiable rewards (RLVR) has improved the reasoning abilities of large language models (LLMs), but most existing approaches rely on sparse outcome-level feedback. This sparsity creates a credit assig…
arXiv:2605.27485v1 Announce Type: cross Abstract: Formal verification offers a path to provably correct software, but writing verified code remains expensive enough that the technique is rarely used in production. Recent large language models can accelerate this work, and recent …
arXiv:2602.10538v3 Announce Type: replace-cross Abstract: Agentic theorem provers combine a reasoning model, retrieval, search, and a proof assistant verifier, yet it remains unclear which components actually improve finite-budget proof success and why they help on real mathemati…
arXiv cs.AI
TIER_1English(EN)·Alessandro Sosso, Akhil Arora, Bas Spitters·
arXiv:2605.23772v1 Announce Type: new Abstract: Agentic systems have recently emerged as state-of-the-art approaches for automated theorem proving in formal mathematics. To assess how far these capabilities extend to program verification, we evaluate Claude Code in an agentic pro…
arXiv cs.AI
TIER_1English(EN)·Benjamin Breen, Marco Del Tredici, Jacob McCarran, Javier Aspuru Mijares, Weichen Winston Yin, Kfir Sulimany, Jacob M. Taylor, Frank H. L. Koppens, Dirk Englund·
arXiv:2510.12787v4 Announce Type: replace Abstract: We present Ax-Prover, a multi-agent system for automated theorem proving in Lean that can solve problems across diverse scientific domains and operate either autonomously or collaboratively with human experts. To achieve this, A…
Agentic systems have recently emerged as state-of-the-art approaches for automated theorem proving in formal mathematics. To assess how far these capabilities extend to program verification, we evaluate Claude Code in an agentic proving framework on CLEVER, a Lean 4 benchmark for…
arXiv cs.CL
TIER_1English(EN)·Riyaz Ahuja, Jeremy Avigad, Prasad Tetali, Sean Welleck·
arXiv:2410.04753v2 Announce Type: replace-cross Abstract: Large language models (LLMs) have been used to generate formal proofs of mathematical theorems in proofs assistants such as Lean. However, we often want to optimize a formal proof with respect to various criteria, dependin…
