AI Paper Atlas

Research map / Large language models / Agents, code and math

SorryDB: Can AI Provers Complete Real-World Lean Theorems?

Austin Letson, Leopoldo Sarra, Auguste Poiroux, Oliver Dressler, Paul Lezeau, Dhyan Aranha, Frederick Pu, Aaron Hill, Miguel Corredera Hidalgo, Julian Berman, George Tsoukalas, Lenny Taelman

ICML 2026 · Large language models › Agents, code and math · approach: Retrieval-augmented formal theorem proving · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add compiler feedback self-correction (16 iterations)Claude Opus 4.5 · SorryDB-2601success rate: 7.8 % -> 27.1 %Improved
add compiler feedback self-correction (16 iterations)Gemini Flash 3 · SorryDB-2601success rate: 10.8 % -> 27.9 %Improved
add LeanSearch tool agent on top of self-correctionGemini Flash 3 · SorryDB-2601success rate: 27.9% -> 30.3%Improved
increase samples from pass@1 to pass@32Gemini Flash 3 · SorryDB-2601success rate: 10.8 % -> 20.5 %Improved
use general-purpose LLM instead of specialized proverGoedel Prover V2 32B vs Gemini Flash 3 · SorryDB-2601pass@1: 10.8 % -> 2.7%Hurt
combine all proversall provers · SorryDB-2601success rate: 30.3 % -> 35.7 %Improved
use deterministic tactics instead of LLM for some tasksTactics · SorryDB-2601success rate: 8.4 %Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy