MolCap-Arena: A Comprehensive Captioning Benchmark on Language-Enhanced Molecular Property Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Edwards, Carl, Lu, Ziqing, Hajiramezanali, Ehsan, Biancalani, Tommaso, Ji, Heng, Scalia, Gabriele
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929572120887296
author Edwards, Carl
Lu, Ziqing
Hajiramezanali, Ehsan
Biancalani, Tommaso
Ji, Heng
Scalia, Gabriele
author_facet Edwards, Carl
Lu, Ziqing
Hajiramezanali, Ehsan
Biancalani, Tommaso
Ji, Heng
Scalia, Gabriele
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2411_00737
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle MolCap-Arena: A Comprehensive Captioning Benchmark on Language-Enhanced Molecular Property Prediction
Edwards, Carl
Lu, Ziqing
Hajiramezanali, Ehsan
Biancalani, Tommaso
Ji, Heng
Scalia, Gabriele
Computation and Language
Artificial Intelligence
Biomolecules
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.
title MolCap-Arena: A Comprehensive Captioning Benchmark on Language-Enhanced Molecular Property Prediction
topic Computation and Language
Artificial Intelligence
Biomolecules
url https://arxiv.org/abs/2411.00737