Chem4DLLM: 4D Multimodal LLMs for Chemical Dynamics Understanding

2026-03-12

AI Summary

This work addresses the limitation of existing chemical understanding methods, which rely on static molecular representations and struggle to capture dynamic processes such as bond breaking and conformational changes. To bridge this gap, we introduce Chemical Dynamics Understanding (ChemDU), a novel task that translates 4D molecular trajectories into interpretable natural language descriptions of key dynamic events in both gas-phase and catalytic reactions. We present Chem4DBench, the first benchmark dataset pairing 4D molecular trajectories with expert-written explanations, and propose Chem4DLLM, a unified architecture integrating equivariant graph neural networks with large language models to jointly model molecular geometry, rotational dynamics, and linguistic semantics. Experimental results demonstrate that our model generates coherent, mechanistically plausible explanations of dynamic chemical processes, showing strong performance and promising potential for this new task.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: (Large) Language Models
• Knowledge Representation and Reasoning: Description Logics

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding

Abstract

Existing chemical understanding tasks primarily rely on static molecular representations, limiting their ability to model inherently dynamic phenomena such as bond breaking or conformational changes, which are essential for a chemist to understand chemical reactions. To address this gap, we introduce Chemical Dynamics Understanding (ChemDU), a new task that translates 4D molecular trajectories into interpretable natural-language explanations. ChemDU focuses on fundamental dynamic scenarios, including gas-phase and catalytic reactions, and requires models to reason about key events along molecular trajectories, such as bond formation and dissociation, and to generate coherent, mechanistically grounded narratives. To benchmark this capability, we construct Chem4DBench, the first dataset pairing 4D molecular trajectories with expert-authored explanations across these settings. We further propose Chem4DLLM, a unified model that integrates an equivariant graph encoder with a pretrained large language model to explicitly capture molecular geometry and rotational dynamics. We hope that ChemDU, together with Chem4DBench and Chem4DLLM, will stimulate further research in dynamic chemical understanding and multimodal scientific reasoning.

Problem

• chemical dynamics
• molecular trajectories
• bond dissociation
• conformational changes
• dynamic chemical understanding

Innovation

• Chemical Dynamics Understanding
• 4D Molecular Trajectories
• Multimodal LLM
• Equivariant Graph Encoder
• Chem4DBench