StochBench: Lean 4 Benchmark for Stochastic Processes
September 7, 2026
StochBench introduces 450 graduate-level stochastic processes problems in Lean 4 to evaluate formal theorem proving. An Opus 4.8-based agent achieved a 34.9% proof rate, addressing a gap in current math-centric LLM benchmarks.
HOW THIS AFFECTS YOU
●
researcherYou can use this specialized dataset to evaluate LLM capabilities in applied mathematical formalization beyond competition math.