RxnExBench: Benchmarking Vision-Language Models for Structured Chemical Reaction Extraction

Evaluation of 9 state-of-the-art VLMs shows that performance degrades substantially as schema complexity and reasoning demands increase. It comes from a paper posted to arXiv on 27 October 2026. Screen Tables and Substrate Scopes are information-dense figures widely used in synthetic chemistry literature, yet automated extraction of structured reaction data from them remains challenging. We introduce RxnExBench, a multimodal benchmark for fine-grained reaction-level evaluation. Built from high-impact chemistry journals, it contains 3,565 expert-annotated reactions and 2,361 molecular structures. The dataset adopts a unified schema with 15 reaction roles and 26 core fields, supporting downstream AI-for-Chemistry tasks such as retrosynthetic planning, yield prediction, and reaction condition recommendation. RxnExBench further provides a reasoning-driven difficulty taxonomy, a modular annotation pipeline, and the Oracle-MolRec protocol, which disentangles molecular recognition from higher-level reasoning. Under the Oracle-MolRec setting, results further reveal that accurate molecular recognition alone is insufficient, with multimodal integration and chemistry-aware reasoning remaining the primary bottlenecks. We release all data and code to facilitate future research in chemistry-aware multimodal understanding and AI-for-Chemistry dataset construction. The dataset and code will be released at https://github.com/opendatalab/RxnExBench.

Why it counts

Evaluation of 9 state-of-the-art VLMs shows that performance degrades substantially as schema complexity and reasoning demands increase.

Sources

Primary source: primary source

What is not known

This brief does not claim independent replication. Claims that appear only on X and not in the primary source stay unknown.