Reasoning-Enhanced Large Language Models for Molecular Property Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Zhuang, Jiaxi, Shi, Yaorui, Hou, Jue, He, Yunong, Ye, Mingwei, Xu, Mingjun, Su, Yuming, Zhang, Linfeng, Qian, Ying, Ke, Guolin, Cai, Hengxing
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911216069246976
author Zhuang, Jiaxi
Shi, Yaorui
Hou, Jue
He, Yunong
Ye, Mingwei
Xu, Mingjun
Su, Yuming
Zhang, Linfeng
Qian, Ying
Zhang, Linfeng
Ke, Guolin
Cai, Hengxing
author_facet Zhuang, Jiaxi
Shi, Yaorui
Hou, Jue
He, Yunong
Ye, Mingwei
Xu, Mingjun
Su, Yuming
Zhang, Linfeng
Qian, Ying
Zhang, Linfeng
Ke, Guolin
Cai, Hengxing
contents Molecular property prediction is crucial for drug discovery and materials science, yet existing approaches suffer from limited interpretability, poor cross-task generalization, and lack of chemical reasoning capabilities. Traditional machine learning models struggle with task transferability, while specialized molecular language models provide little insight into their decision-making processes. To address these limitations, we propose \textbf{MPPReasoner}, a multimodal large language model that incorporates chemical reasoning for molecular property prediction. Our approach, built upon Qwen2.5-VL-7B-Instruct, integrates molecular images with SMILES strings to enable comprehensive molecular understanding. We develop a two-stage training strategy: supervised fine-tuning (SFT) using 16,000 high-quality reasoning trajectories generated through expert knowledge and multiple teacher models, followed by Reinforcement Learning from Principle-Guided Rewards (RLPGR). RLPGR employs verifiable, rule-based rewards that systematically evaluate chemical principle application, molecular structure analysis, and logical consistency through computational verification. Extensive experiments across 8 datasets demonstrate significant performance improvements, with MPPReasoner outperforming the best baselines by 7.91\% and 4.53\% on in-distribution and out-of-distribution tasks respectively. MPPReasoner exhibits exceptional cross-task generalization and generates chemically sound reasoning paths that provide valuable insights into molecular property analysis, substantially enhancing both interpretability and practical utility for chemists. Code is available at https://anonymous.4open.science/r/MPPReasoner-12687.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10248
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Reasoning-Enhanced Large Language Models for Molecular Property Prediction
Zhuang, Jiaxi
Shi, Yaorui
Hou, Jue
He, Yunong
Ye, Mingwei
Xu, Mingjun
Su, Yuming
Zhang, Linfeng
Qian, Ying
Zhang, Linfeng
Ke, Guolin
Cai, Hengxing
Machine Learning
Artificial Intelligence
Molecular property prediction is crucial for drug discovery and materials science, yet existing approaches suffer from limited interpretability, poor cross-task generalization, and lack of chemical reasoning capabilities. Traditional machine learning models struggle with task transferability, while specialized molecular language models provide little insight into their decision-making processes. To address these limitations, we propose \textbf{MPPReasoner}, a multimodal large language model that incorporates chemical reasoning for molecular property prediction. Our approach, built upon Qwen2.5-VL-7B-Instruct, integrates molecular images with SMILES strings to enable comprehensive molecular understanding. We develop a two-stage training strategy: supervised fine-tuning (SFT) using 16,000 high-quality reasoning trajectories generated through expert knowledge and multiple teacher models, followed by Reinforcement Learning from Principle-Guided Rewards (RLPGR). RLPGR employs verifiable, rule-based rewards that systematically evaluate chemical principle application, molecular structure analysis, and logical consistency through computational verification. Extensive experiments across 8 datasets demonstrate significant performance improvements, with MPPReasoner outperforming the best baselines by 7.91\% and 4.53\% on in-distribution and out-of-distribution tasks respectively. MPPReasoner exhibits exceptional cross-task generalization and generates chemically sound reasoning paths that provide valuable insights into molecular property analysis, substantially enhancing both interpretability and practical utility for chemists. Code is available at https://anonymous.4open.science/r/MPPReasoner-12687.
title Reasoning-Enhanced Large Language Models for Molecular Property Prediction
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2510.10248