AI Paper Atlas

Research map / Continual learning / Domain and prompt-based

OptProver: Bridging Olympiad and Optimization through Continual Training in Formal Theorem Proving

Chenyi Li, Yanchen Nie, Zhenyu Ming, Gong Zhang, Kun Yuan, Zaiwen Wen

ICML 2026 · Continual learning › Domain and prompt-based · approach: Domain-incremental learning with memory · 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 two rounds of expert iteration to BFS-Prover-V2BFS-Prover-V2-7B · OptBenchPass@1 average: 13.25% -> 28.75%Improved
add two rounds of expert iteration to BFS-Prover-V2BFS-Prover-V2-7B · OptBenchPass@32 average: 32.00% -> 50.00%Improved
add DPO after expert iteration (ablation)OptProver (Only EI) · OptBenchPass@32 average: 50.00% -> 52.00%Improved
use UAPO instead of DPO (ablation)OptProver (EI) · OptBenchPass@32 average: 52.00% -> 53.50%Improved
use PW-DPO instead of DPO (ablation)OptProver (EI) · OptBenchPass@32 average: 52.00% -> 54.25%Improved
combine utility-aware and perplexity-weighted (PW-UAPO) (ablation)OptProver (EI) · OptBenchPass@32 average: 54.25% -> 55.25%Improved
use step-level proving instead of whole-proof generation7B provers · OptBenchaverage Pass@32: 19.75% -> 32.00%Improved
set PW-DPO tau to median 1.08 instead of 1.00 (ablation)PW-DPO · OptBenchPass@1: 36.00% -> 36.25%Improved
raise PW-DPO tau to 2.42 (ablation)PW-DPO · OptBenchPass@1: 36.25% -> 32.75%Hurt
apply OptProver training to base modelBFS-Prover-V2-7B · ProofNetproblems solved at Pass@1: 21 -> 31Improved
apply OptProver training to base modelBFS-Prover-V2-7B · ProofNetproblems solved at Pass@32: 42 -> 49Improved
apply OptProver training to base modelBFS-Prover-V2-7B · MiniF2FPass@32No clear effect

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