MolCap-Arena: A Comprehensive Captioning Benchmark on Language-Enhanced Molecular Property Prediction
This work addresses the need for a standardized benchmark to evaluate LLM-driven insights in biomolecular modeling, which is incremental as it builds on existing interdisciplinary research.
The study tackled the problem of assessing how large language models (LLMs) can improve molecular property prediction by introducing MolCap-Arena, a comprehensive benchmark evaluating over twenty LLMs across diverse tasks, and found that LLM-extracted knowledge enhances state-of-the-art molecular representations with variations based on model, prompt, and dataset.
Bridging biomolecular modeling with natural language information, particularly through large language models (LLMs), has recently emerged as a promising interdisciplinary research area. LLMs, having been trained on large corpora of scientific documents, demonstrate significant potential in understanding and reasoning about biomolecules by providing enriched contextual and domain knowledge. However, the extent to which LLM-driven insights can improve performance on complex predictive tasks (e.g., toxicity) remains unclear. Further, the extent to which relevant knowledge can be extracted from LLMs also remains unknown. In this study, we present Molecule Caption Arena: the first comprehensive benchmark of LLM-augmented molecular property prediction. We evaluate over twenty LLMs, including both general-purpose and domain-specific molecule captioners, across diverse prediction tasks. To this goal, we introduce a novel, battle-based rating system. Our findings confirm the ability of LLM-extracted knowledge to enhance state-of-the-art molecular representations, with notable model-, prompt-, and dataset-specific variations. Code, resources, and data are available at github.com/Genentech/molcap-arena.