StochBench: A Domain-Specific Benchmark for Stochastic Processes in Lean

  • 类型:arxiv
  • 标识:2609.09264
  • 链接:https://arxiv.org/abs/2609.09264
  • 主分类:evaluation
  • 形态:benchmark
  • 被引:1
  • 被引来源:Semantic Scholar
  • S2被引:1
  • OpenAlex被引:0
  • 影响力被引:0
  • TLDR:StochBench is introduced, a Lean 4 benchmark of 450 graduate stochastic-processes problems at varying abstraction levels, each paired with its natural-language source that better represents domain-specific applied mathematics while remaining challenging for advanced provers.
  • OpenAlex ID:W7212133990
  • OpenAlex DOI:10.48550/arxiv.2609.09264
  • DOI:10.48550/arxiv.2609.09264
  • DOI来源:OpenAlex
  • 开放获取:green
  • 开放获取链接:https://doi.org/10.48550/arxiv.2609.09264
  • OpenAlex更新:2026-09-13
  • 待LLM分类:否
  • 标题中文:StochBench:面向 Lean 中随机过程的领域专用基准
  • TLDR中文:本文介绍 StochBench,一个基于 Lean 4 的基准,包含 450 道覆盖不同抽象层次的研究生随机过程问题,每道题配有其自然语言来源,更能代表领域特定的应用数学,同时对高级证明器仍具挑战。
  • 来源文件:
  • /inbox/tom/_candidates/2026-09-10-agent-rag-longcontext-candidates.json
  • /inbox/tom/_candidates/2026-09-11-agent-rag-longcontext-candidates.json
  • [OpenAlex backfill]
  • [S2 enrich]