AI Paper Atlas

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

ProSearch: Benchmarking Multi-Constraint Protocol Retrieval in Experimental Science

Wenliang Liang, Zhiyuan Ning, Haolin Chen, Yuanchun Zhou, Wenjuan Cui, Yi Du

NeurIPS 2026 · Large language models › Agents, code and math · approach: LLMs for scientific discovery · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Introduces ProSearch, a benchmark for evaluating retrievers on multi-constraint experimental protocol retrieval in science.

Problem

Retrieval systems supporting experimental design and method reproduction must find protocols satisfying interdependent constraints (materials, operating conditions), but systematic evaluation of retrievers in such multi-constraint settings is lacking.

Approach

A benchmark dedicated to multi-constraint protocol retrieval in experimental science, providing a systematic evaluation setting for retrievers.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
rerank top-100 with a cross-encoder or LLM reranker (Qwen3-Reranker)Qwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.474Improved
add DIVER-QExpand query expansionQwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.317 (+0.027)Improved
add reasoning-augmented query rewritingQwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.310 (+0.020)Improved
decompose the query into sub-queriesQwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.232Hurt
use HyDE hypothetical documents as the queryQwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.272Hurt
rewrite query as structured queryQwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.246Hurt
TongSearch-QR query rewritingQwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.285No clear effect
use scientific-domain pretrained retrievers instead of general dense retrieversdense retrievers · ProSearch avgnDCG@10: best scientific 0.158 vs best general 0.290Hurt
use reasoning-oriented retrievers instead of general dense retrieversdense retrievers · ProSearch avgnDCG@10: best reasoning 0.255 vs 0.290No clear effect
expand the corpus from method paragraphs to all body paragraphs (ablation)Qwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.206Hurt
add explicit structured constraint matching over extracted records (ablation)Qwen3-Embedding-8B · ProSearch avgnDCG@10: 0.290 -> 0.774Improved
break structured-match ties with dense similarity (ablation)Structured Diagnostic · ProSearch avgnDCG@10: 0.774 -> 0.814Improved

Experimental setup

Benchmarks: ProSearch

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