arXiv cs.CL· Yicheng Tao, Yiqun Wang, Xiangchen Song, Xin Luo, Kai Liu, Yuanhao Huang, Jie Liu·· 2 天前AI 评分22
GRASP：面向半结构化知识库的规划引导图检索与自适应融合重排序
GRASP: Plan-Guided Graph Retrieval with Adaptive Fusion and Reranking on Semi-Structured Knowledge Bases
AI 导读
GRASP 是一种混合检索器，通过小型 LLM 规划器将查询转为可执行图计划，融合图排序与文本排序，并用微调重排序器重排候选结果。在三个 STaRK 基准上，GRASP 所有指标均取得最优，平均 Hit@1 较最强基线提升 18.6 分，单次查询耗时仅 2.47 秒。
正文 · 原文
Abstract:Retrieval over semi-structured knowledge bases (SKBs) must handle both textual descriptions and relational constraints. Text-based retrievers capture the overall semantics of a query but struggle with multi-hop relations, whereas graph-based retrievers follow relational paths precisely but miss the broader meaning. We propose GRASP, a hybrid retriever that learns to fuse the two. A small LLM planner translates the query into an executable graph plan in a single call, its graph ranking is fused with a text ranking, and a fine-tuned reranker reorders the fused candidates. Instead of training the planner to be accurate on its own, we optimize it with reinforcement learning to maximize the gain its graph ranking brings to the fused ranking. On three STaRK benchmarks, GRASP achieves the best results on all metrics, improving the average Hit@1 by 18.6 points over the strongest baseline, while taking only 2.47 seconds per query.
| Subjects: | Information Retrieval (cs.IR); Computation and Language (cs.CL); Machine Learning (cs.LG) |
| Cite as: | arXiv:2605.30237 [cs.IR] |
| (or arXiv:2605.30237v3 [cs.IR] for this version) | |
| https://doi.org/10.48550/arXiv.2605.30237 arXiv-issued DOI via DataCite |
Submission history
From: Yicheng Tao [view email]
[v1]
Thu, 28 May 2026 17:07:32 UTC (7,041 KB)
[v2]
Sun, 31 May 2026 00:43:51 UTC (7,041 KB)
[v3]
Thu, 15 Oct 2026 17:59:16 UTC (966 KB)
来源：arXiv cs.CL · arxiv.org