arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2604.08952cs.CLcs.IR
收藏

MAB-DQA: 通过多臂老虎机解决文档问答中的查询方面重要性

MAB-DQA: Addressing Query Aspect Importance in Document Question Answering with Multi-Armed Bandits

Nanjing University of Science and Technology(南京理工大学);Singapore Management University(新加坡国立大学);Nanjing Pami Intelligent Technology Co., Ltd.(南京帕米智能科技有限公司);University of Wisconsin - Madison(威斯康星大学麦迪逊分校);Nanjing Forestry University(南京林业大学)

Yixin Xiang, Yunshan Ma, Xiaoyu Du, Yibing Chen, Yanxin Zhang, Jinhui Tang

AI总结:

本文提出MAB-DQA框架,通过多臂老虎机模型解决文档问答中查询方面的重要性问题,通过分解查询为方面感知的子查询并动态分配检索预算,提升文档理解性能。

摘要:

文档问答(DQA)涉及根据用户查询从文档中生成答案,是文档理解的关键任务。该任务需要解释视觉布局,促使近期研究采用多模态检索增强生成(RAG)来处理页面图像进行答案生成。然而,在多模态RAG中,视觉DQA难以有效利用大量图像,因为检索阶段通常只保留少数候选页面(例如Top-4),导致信息丰富但视觉不显著的内容被忽视,而选择常见但信息少的页面。为了解决这个问题,我们提出了一种基于多臂老虎机的DQA框架(MAB-DQA),以显式建模查询中多个隐含方面的不同重要性。具体来说,MAB-DQA将查询分解为方面感知的子查询,并为每个方面检索特定的候选集。它将每个子查询视为一个臂,并利用少量代表性页面的初步推理结果作为奖励信号来估计方面效用。在探索-利用策略的指导下,MAB-DQA动态重新分配检索预算,以指向高价值的方面。通过最有信息量的页面及其相关性,MAB-DQA生成预期结果。在四个基准测试中,MAB-DQA在最先进方法上平均提高了5%-18%,一致提升了文档理解能力。代码可在https://github.com/ElephantOH/MAB-DQA获取。

英文摘要:

Document Question Answering (DQA) involves generating answers from a document based on a user's query, representing a key task in document understanding. This task requires interpreting visual layouts, which has prompted recent studies to adopt multimodal Retrieval-Augmented Generation (RAG) that processes page images for answer generation. However, in multimodal RAG, visual DQA struggles to utilize a large number of images effectively, as the retrieval stage often retains only a few candidate pages (e.g., Top-4), causing informative but less visually salient content to be overlooked in favor of common yet low-information pages. To address this issue, we propose a Multi-Armed Bandit-based DQA framework (MAB-DQA) to explicitly model the varying importance of multiple implicit aspects in a query. Specifically, MAB-DQA decomposes a query into aspect-aware subqueries and retrieves an aspect-specific candidate set for each. It treats each subquery as an arm and uses preliminary reasoning results from a small number of representative pages as reward signals to estimate aspect utility. Guided by an exploration-exploitation policy, MAB-DQA dynamically reallocates retrieval budgets toward high-value aspects. With the most informative pages and their correlations, MAB-DQA generates the expected results. On four benchmarks, MAB-DQA shows an average improvement of 5%-18% over the state-of-the-art method, consistently enhancing document understanding. Codes are available at https://github.com/ElephantOH/MAB-DQA.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑