MolGround: A Benchmark for Molecular Grounding
This work addresses the problem of molecular grounding for researchers in AI for Science, cheminformatics, and NLP, though it appears incremental as it builds on existing conventions and techniques.
The paper tackles the gap in linking molecular concepts to specific structural components by proposing MolGround, a benchmark for molecular grounding that evaluates models' referential abilities. They constructed the largest molecular understanding benchmark with 117k QA pairs and developed a multi-agent prototype that outperforms existing models like GPT-4o, enhancing tasks such as molecular captioning and ATC classification.
Current molecular understanding approaches predominantly focus on the descriptive aspect of human perception, providing broad, topic-level insights. However, the referential aspect -- linking molecular concepts to specific structural components -- remains largely unexplored. To address this gap, we propose a molecular grounding benchmark designed to evaluate a model's referential abilities. We align molecular grounding with established conventions in NLP, cheminformatics, and molecular science, showcasing the potential of NLP techniques to advance molecular understanding within the AI for Science movement. Furthermore, we constructed the largest molecular understanding benchmark to date, comprising 117k QA pairs, and developed a multi-agent grounding prototype as proof of concept. This system outperforms existing models, including GPT-4o, and its grounding outputs have been integrated to enhance traditional tasks such as molecular captioning and ATC (Anatomical, Therapeutic, Chemical) classification.