arXiv cs.AI· Zekai Chen, Hao Zhu, Yu Liu, Xunkai Li, Zejian Kang, Rui Qian, Quanjiang Li, Rong-Hua Li·· 2 小时前AI 评分31
RouteDRAG：面向多模态文档 RAG 的路由与 Agentic 执行控制学习框架
RouteDRAG: Learning to Route and Control Agentic Execution for Multimodal Document RAG
AI 导读
RouteDRAG 是一个学习框架，通过统一文档索引、自适应路径路由、动态 Agent 执行和序列策略学习，控制多模态文档 RAG 中智能体的调用与执行。在两个 benchmark 上，它比仅短路径执行平均提升 3.47 个百分点，同时只将 79.3% 的查询分配给 Agent。它还在两个 benchmark 上超过统一 Agent 执行，且所需的回答调用更少，代码已开源。
正文 · 原文
Abstract:Web-based applications increasingly rely on understanding long multimodal documents, where answering questions often requires combining information across pages, tables, and figures. Agentic retrieval-augmented generation addresses these tasks through iterative retrieval and reasoning. However, Agentic execution is not always necessary, introducing additional computation and potentially incorrect answers. Our empirical analysis reveals that Agentic execution provides query-dependent benefits, while additional operations may improve, preserve, or degrade answer accuracy. These findings motivate controlling both Agent invocation and subsequent execution. To address these challenges, we propose RouteDRAG, a learning-based framework comprising four components. Unified Document Index organizes multimodal content into a graph-backed index and derives query representations from initial retrieval. Adaptive Path Routing preserves exact answering and predicts the relative benefit of Agent over shorter execution paths. Dynamic Agent Execution maintains an evolving workspace and refines Planner proposals to control tool use and termination. Sequential Policy Learning first trains operation control from observed outcomes and then trains path routing using the resulting Agent's performance. Across two benchmarks, RouteDRAG achieves an average accuracy improvement of 3.47 percentage points over shorter-path-only execution while assigning only 79.3% of queries to Agent. Notably, it also surpasses uniform Agent execution on both benchmarks while requiring fewer answering calls. Our code is now available at this https URL.
| Comments: | Our code is now available at this https URL |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.36444 [cs.AI] |
| (or arXiv:2610.36444v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.36444 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Rui Qian [view email]
[v1]
Tue, 27 Oct 2026 00:45:49 UTC (1,222 KB)
来源：arXiv cs.AI · arxiv.org