arXiv cs.AI· Hang Zhang, Qihao Ye, Litong Liu, Yunbei Zhang, Yugyung Lee, Yue Zhou, Bowen Liu, Emily First, Cheng Han, Qiao Zhuang·· 4 小时前AI 评分31
LeanContext 基准：将数学资源与局部形式化上下文解耦的 Lean 整证明生成研究
LeanContext: Disentangling Mathematical Resources from Local Formal Context in Lean Whole-Proof Generation
AI 导读
研究者发布 LeanContext 基准，包含 129,826 道 Lean 整证明生成问题，用于评估检索前提与项目上下文对神经定理证明的价值。基准将选中声明分为数学上下文（MC）与环境上下文（EC），并支持四种对齐输入视图。
正文 · 原文
Abstract:Neural theorem proving increasingly uses retrieved premises and project context to support proof generation. Evaluating these resources requires measuring the advantage of target-selected declarations over comparable alternatives and how this advantage depends on training. We introduce LeanContext, a benchmark of 129,826 Lean whole-proof generation problems, to investigate these questions. We group selected complete preceding mathematical declarations as Mathematical Context (MC) and the remaining selected local source commands as Environment Context (EC). Four aligned input views support crossed training and inference evaluations with shared targets and original Lean verification environments. For both diffusion- and autoregressive-based models, selected declarations outperform comparable alternatives by a larger margin when training includes MC. Final-input audits identify omitted declarations referenced by the target. Prioritizing these dependencies improves Pass@32 by 1.40% and 1.70% for LLaDA and Llama, respectively, on 2,000 targets. The selection experiment operates without access to target reference proofs and holds checkpoints, EC, and inference budgets constant. The code and dataset are available at this https URL.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.08277 [cs.AI] |
| (or arXiv:2611.08277v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.08277 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Hang Zhang [view email]
[v1]
Thu, 5 Nov 2026 16:44:59 UTC (582 KB)
来源：arXiv cs.AI · arxiv.org