arXiv cs.AI· Maya Arseven, Anette Frank, Beni Egressy, Johann Higl, Moritz Plenz·· 9 天前AI 评分23
GLM-RAG：面向基于图的检索增强生成的图语言模型
GLM-RAG: Graph Language Models for Graph-Based Retrieval-Augmented Generation
AI 导读
研究提出基于图语言模型（GLM）的检索器 GLM-RAG，并对比 GLM、GNN 与向量搜索三类检索器在单跳和多跳 RAG 中的表现。微调后的 GLM 检索器跨域泛化更优，在两个多跳基准上达到 SOTA，域内多跳 QA 与既有工作相当，且随参数与子图覆盖提升有良好扩展性。GNN 检索器以高效训练获得更高图覆盖，向量搜索基线则在单跳数据集上表现最佳。
正文 · 原文
Abstract:Retrieval-augmented generation (RAG) over knowledge graphs requires retrievers that can effectively capture both graph structure and semantic information. Recent approaches have explored graph neural network (GNN)-based retrievers to model graph topology in multi-hop reasoning tasks. In parallel, graph language models (GLMs) have emerged as a promising paradigm that integrates graph reasoning and the semantic capabilities of language models. In this work, we introduce a GLM-based retriever and investigate the comparative strengths of GLM-based, GNN-based, and traditional vector-search-based retrievers in single- and multi-hop RAG settings, and with a particular focus on transferability to unseen domains. Our findings suggest that finetuned GLM retrievers generalize better out of domain, achieving SOTA on two multi-hop benchmarks. On in-domain multi-hop QA datasets they remain comparable to prior work, with promising scaling as parameters and subgraph coverage increase. GNN-based retrievers achieve higher graph coverage with an efficient training setup, whereas the vector-search baseline excels at single-hop datasets.
| Comments: | Accepted to AACL-IJCNLP 2026, 9 pages, 19 figures |
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Information Retrieval (cs.IR) |
| ACM classes: | I.2.0; I.2.4; I.2.7 |
| Cite as: | arXiv:2607.28397 [cs.AI] |
| (or arXiv:2607.28397v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.28397 arXiv-issued DOI via DataCite |
Submission history
From: Maya Arseven [view email]
[v1]
Thu, 30 Jul 2026 15:49:34 UTC (572 KB)
[v2]
Thu, 8 Oct 2026 11:33:14 UTC (607 KB)
来源：arXiv cs.AI · arxiv.org