AI Paper Atlas

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

BioProBench: A Corpus and Benchmark for Biological Protocol Reasoning in Autonomous Science

Yuyang Liu, Liuzhenghao Lv, Xiancheng Zhang, Jingya Wang, Li Yuan, Yonghong Tian

ICML 2026 · Large language models › Agents, code and math · approach: LLMs for scientific discovery

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 structured chain-of-thought prompting (REA-ERR vs ERR)QwQ-32b · BioProBench ERRF1: 59.41% -> 71.40%Improved
add zero-shot CoT prompting to protocol generationClaude-3-7-sonnet · BioProBench GENStep Recall: 0.4280 to 0.3918Hurt
add BioProCorpus retrieval (ProAgent) for PQAGemini-2.5-pro-exp · BioProBench PQAaccuracy: 15 points to 85.08%Improved
add BioProCorpus retrieval (ProAgent) for ERRGemini-2.5-pro-exp · BioProBench ERRF1: 59.27% -> 81.9%Improved
add BioProCorpus retrieval (ProAgent) for GENClaude-3-7-sonnet · BioProBench GENStep Recall: 42.8% to 62.24%Improved
add BioProBench RAG module on external benchmarkProAgent base model · LAB-Bench protocol subset (108 questions)accuracy: 56% -> 67%Improved
perturb numerical values in PQA context (ablation)PQA evaluation · BioProBench PQAaccuracyNo clear effect
add one-shot CoT exemplar instead of zero-shotGEN prompting · BioProBench GENmixed metricsMixed

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