cs.CVNov 13, 2026

DoubleTake: Focused Verification of Visual Evidence for Long-Document Understanding

Authors: Kejun Xue, Ang Li, Jinghua Tang, Xincheng Wang, Yongbin Gao, Yanghua Xiao, Zhijun Fang

Abstract

In long-document question answering, a vision-language model can answer incorrectly even when the retrieved pages contain sufficient evidence. Rereading those pages with the original question gives the model little guidance about which fact to check. We introduce DoubleTake, a training-free framework for focused verification of visual evidence. An initial reading produces a candidate answer and a separate evidence record that guides retrieval for a second reading. A planner compares the two candidates and their supporting evidence to formulate a factual question. A verifier answers it from the source images without seeing either candidate or evidence record, and a reasoner uses this observation to resolve the original question. With a frozen Qwen3-VL-8B-Instruct backbone, DoubleTake achieves the highest overall accuracy among the evaluated methods on each of four benchmarks. With the same pages and number of model calls, asking a focused factual question yields higher accuracy on all four benchmarks than asking the original question again. Our code is available at https://github.com/small-code-cat/double_take.

Figures & tables

Explore similar work

CardsList
  1. VELD-QA: A Large-Scale Dataset for Visual Evidence Grounding and Question Answering in Long Documents

    Oct 14, 2026Zhivar Sourati, Marianne Menglin Liu, Rongguang Wang +12Long-Document Visual Question AnsweringDocument QA

  2. DocScope: Benchmarking Verifiable Reasoning for Trustworthy Long-Document Understanding

    May 9, 2026Xiang Feng, Jiawei Zhou, Zhangfeng Huang +6Multimodal UnderstandingLLM Answer Verification

  3. Doc-V*:Coarse-to-Fine Interactive Visual Reasoning for Multi-Page Document VQA

    Apr 15, 2026Yuanlei Zheng, Pei Fu, Hang Li +9Document QALong-Document Visual Question Answering