Bridging Molecular Graphs and Large Language Models
This addresses the challenge of integrating molecular graph data into LLMs for researchers in computational chemistry or drug discovery, representing an incremental improvement by adapting existing LLM frameworks.
The paper tackled the problem of limited graph data processing in Large Language Models (LLMs) by proposing Graph2Token, which aligns graph tokens to LLM tokens without fine-tuning, enabling molecular few-shot learning and achieving effectiveness in classification and regression tasks.
While Large Language Models (LLMs) have shown exceptional generalization capabilities, their ability to process graph data, such as molecular structures, remains limited. To bridge this gap, this paper proposes Graph2Token, an efficient solution that aligns graph tokens to LLM tokens. The key idea is to represent a graph token with the LLM token vocabulary, without fine-tuning the LLM backbone. To achieve this goal, we first construct a molecule-text paired dataset from multisources, including CHEBI and HMDB, to train a graph structure encoder, which reduces the distance between graphs and texts representations in the feature space. Then, we propose a novel alignment strategy that associates a graph token with LLM tokens. To further unleash the potential of LLMs, we collect molecular IUPAC name identifiers, which are incorporated into the LLM prompts. By aligning molecular graphs as special tokens, we can activate LLM generalization ability to molecular few-shot learning. Extensive experiments on molecular classification and regression tasks demonstrate the effectiveness of our proposed Graph2Token.