LLM and GNN are Complementary: Distilling LLM for Multimodal Graph Learning
This addresses the challenge of leveraging multimodal molecular data for better predictions in chemistry or drug discovery, representing an incremental advancement by combining existing LLM and GNN techniques.
The paper tackles the problem of molecular property prediction by integrating multimodal data (textual and visual) with graph structures, using a framework that distills knowledge from Large Language Models into a Multilayer Perceptron, resulting in notable improvements in accuracy and efficiency.
Recent progress in Graph Neural Networks (GNNs) has greatly enhanced the ability to model complex molecular structures for predicting properties. Nevertheless, molecular data encompasses more than just graph structures, including textual and visual information that GNNs do not handle well. To bridge this gap, we present an innovative framework that utilizes multimodal molecular data to extract insights from Large Language Models (LLMs). We introduce GALLON (Graph Learning from Large Language Model Distillation), a framework that synergizes the capabilities of LLMs and GNNs by distilling multimodal knowledge into a unified Multilayer Perceptron (MLP). This method integrates the rich textual and visual data of molecules with the structural analysis power of GNNs. Extensive experiments reveal that our distilled MLP model notably improves the accuracy and efficiency of molecular property predictions.