GPR: Empowering Generation with Graph-Pretrained Retriever
This addresses the need for more effective graph-specific retrievers in graph retrieval-augmented generation, though it appears incremental as it builds on existing retrieval-augmented generation methods.
The paper tackles the problem of domain misalignment and structure ignorance in graph retrieval-augmented generation by proposing GPR, a graph-based retriever pretrained directly on knowledge graphs, which consistently improves retrieval quality and downstream generation across multiple datasets, LLM backbones, and baselines.
Graph retrieval-augmented generation (GRAG) places high demands on graph-specific retrievers. However, existing retrievers often rely on language models pretrained on plain text, limiting their effectiveness due to domain misalignment and structure ignorance. To address these challenges, we propose GPR, a graph-based retriever pretrained directly on knowledge graphs. GPR aligns natural language questions with relevant subgraphs through LLM-guided graph augmentation and employs a structure-aware objective to learn fine-grained retrieval strategies. Experiments on two datasets, three LLM backbones, and five baselines show that GPR consistently improves both retrieval quality and downstream generation, demonstrating its effectiveness as a robust retrieval solution for GRAG.