cs.AIJul 27, 2025

Building verifiable databases at scale from the chemical literature via a multi-agent system

Authors: Yufan Chen, Ching Ting Leung, Bowen Yu, Haifan Zhou, Jianwei Sun, Yong Huang, Linyan Li, Hao Chen, 

Organizations: Department of Chemical and Biological Engineering, The Hong Kong University of Science and Technology, Hong Kong SAR, China · Department of Chemistry, The Hong Kong University of Science and Technology, Hong Kong SAR, China · Department of Data Science, City University of Hong Kong, Hong Kong SAR, China · Department of Computer Science and Engineering, The Hong Kong University of Science and Technology, Hong Kong SAR, China

Abstract

Accelerating chemical research with artificial intelligence requires high-quality databases. However, many reactions exist only as graphics drawn for human readers, which machines must read exactly. Specialist software is precise but rigid, whereas multimodal language models are flexible yet often write incorrect structures; neither flags errors. We present ChemEAGLE, a multi-agent system where tools supply structures and a language model extends their reach through judgments that programs apply. On 424 published graphics, its F1 score for exactly correct reactions was 82.2%, against 47.7% for a tool-equipped generalist agent, and calibrated confidence limited review to 35% of reactions at 96.6% accuracy. Without review, it built a 60,103-reaction photochemistry database whose records trained models to rank conditions and predict products. Verifiable databases can thus be built at scale.

Figures & tables

Explore similar work

CardsList
  1. RxnExBench: Benchmarking Vision-Language Models for Structured Chemical Reaction Extraction from Scientific Literature

    Oct 23, 2026Jiaxing Sun, Junyuan Gao, Ruikang Zhang +16Multimodal DatasetDocument Information Extraction

  2. Agentic generation of verifiable rules for deterministic, self-expanding reaction classification

    Jul 1, 2026Daniel Armstrong, Maarten Dobbelaere, Valentas Olikauskas +4Rule-Based Classification