Toward Graph-Tokenizing Large Language Models with Reconstructive Graph Instruction Tuning

Researchers have developed RGLM, a new approach to improve how large language models understand and process graph data by incorporating explicit graph supervision alongside text instructions. The method addresses limitations in existing Graph-Tokenizing LLMs that rely too heavily on text supervision, leading to underutilization of graph context.

• →Current Graph-Tokenizing LLMs suffer from text-dominant bias due to relying solely on text supervision from language instructions.
• →RGLM introduces reconstructive graph instruction tuning to explicitly incorporate graph supervision and improve graph-text alignment.
• →The approach includes three variants: RGLM-Decoder, RGLM-Similarizer, and RGLM-Denoiser operating from different perspectives.
• →Information-theoretic analysis shows the alignment objective is bounded by mutual information between input graphs and hidden representations.
• →Extensive experiments validate RGLM's effectiveness across various benchmarks and task scenarios.