提出可自解释的文档视觉问答框架，让模型推理过程可查可验。
Towards Self-Explainable Document Visual Question Answering with Chain-of-Explanation Predictions
- 分三步推理：找相关证据→定位答案区域→仅从该区域解码答案
- 在PFL-DocVQA上提升ANLS 12%，达到当前最佳可解释性能
- 适合需要透明决策过程的文档分析场景，如法律、医疗
文档视觉问答（DocVQA）要求视觉-语言模型不仅判断文档中哪些信息与问题相关，还需定位答案在页面的具体位置。现有模型将相关证据与答案定位混杂，且多为黑箱，难以验证预测如何依赖视觉证据。本文提出CoExVQA框架，通过链式解释设计实现可自解释的文档视觉问答。该框架先识别问题相关的证据，再显式定位答案区域，最后仅从该区域解码答案。基于此链式推理的预测可直接跨模态检验和验证推理过程。实验证明，限制解码仅在有依据的区域进行，在PFL-DocVQA数据集上实现当前最优可解释性能，相比现有可解释基线提升ANLS 12%，同时提供透明且可验证的预测结果。
原文摘要 · Abstract (English)
Document Visual Question Answering (DocVQA) requires vision-language models to reason not only about what information in a document is relevant to a question, but also where the answer is grounded on the page. Existing DocVQA models entangle question-relevant evidence and answer localization and operate largely as black boxes, offering limited means to verify how predictions depend on visual evidence. We propose CoExVQA, a self-explainable DocVQA framework with a grounded reasoning process through a chain-of-explanation design. CoExVQA first identifies question-relevant evidence, then explicitly localizes the answer region, and finally decodes the answer exclusively from the grounded region. Prediction via CoExVQA's chain-of-explanation enables direct inspection and verification of the reasoning process across modalities. Empirical results show that restricting decoding to grounded evidence achieves SotA explainable DocVQA performance on PFL-DocVQA, improving ANLS by 12% over the current explainable baselines while providing transparent and verifiable predictions.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。