AVIR: Adaptive Visual In-Document Retrieval for Efficient Multi-Page Document Question Answering

2026-01-17 · Zongmin Li, Yachuan Li, Lei Kang, Dimosthenis Karatzas, Wenkang Ma arxiv

Multi-page Document Visual Question Answering (MP-DocVQA) remains challenging because long documents not only strain computational resources but also reduce the effectiveness of the attention mechanism in large vision-language models (LVLMs). We tackle these issues with an Adaptive Visual In-document Retrieval (AVIR) framework. A lightweight retrieval model first scores each page for question relevance. Pages are then clustered according to the score distribution to adaptively select relevant content. The clustered pages are screened again by Top-K to keep the context compact. However, for short documents, clustering reliability decreases, so we use a relevance probability threshold to select pages. The selected pages alone are fed to a frozen LVLM for answer generation, eliminating the need for model fine-tuning. The proposed AVIR framework reduces the average page count required for question answering by 70%, while achieving an ANLS of 84.58% on the MP-DocVQA dataset-surpassing previous methods with significantly lower computational cost. The effectiveness of the proposed AVIR is also verified on the SlideVQA and DUDE benchmarks. The code is available at https://github.com/Li-yachuan/AVIR.

📄 PDF Abstract BibTeX arXiv:2601.11976

Code (0)

등록된 구현이 없습니다.

Tasks

Visual Question Answering Answer Generation

NaviRAG: Towards Active Knowledge Navigation for Retrieval-Augmented Generation

2026-04-14 · Jihao Dai, Dingjun Wu, Yuxuan Chen, Zheni Zeng 외 arxiv

Retrieval-augmented generation (RAG) typically relies on a flat retrieval paradigm that maps queries directly to static, isolated text segments. This approach struggles with more complex tasks that require the conditiona…

DocPruner: A Storage-Efficient Framework for Multi-Vector Visual Document Retrieval via Adaptive Patch-Level Embedding Pruning

2025-09-28 · Yibo Yan, Guangwei Xu, Xin Zou, Shuliang Liu 외 arxiv

Visual Document Retrieval (VDR), the task of retrieving visually-rich document pages using queries that combine visual and textual cues, is crucial for numerous real-world applications. Recent state-of-the-art methods le…

MARA: A Multimodal Adaptive Retrieval-Augmented Framework for Document Question Answering

2026-02-01 · Hui Wu, Haoquan Zhai, Yuchen Li, Hengyi Cai 외 arxiv

Retrieval-based multimodal document QA aims to identify and integrate relevant information from visually rich documents with complex multimodal structures. While retrieval-augmented generation (RAG) has shown strong perf…

Question Answering

LightSTAR: Efficient Visual Document Retrieval via Lightweight Selection with Vision-Adaptive Refinement

2026-06-22 · Tongkun Guan, Haocheng Wang, Wei Shen, Xiaokang Yang arxiv

Visual document retrieval requires rapidly locating relevant pages from large multi-modal corpora in response to user queries. While recent methods powered by Multi-modal Large Language Models (MLLMs) show competitive ac…

AutothinkRAG: Complexity-Aware Control of Retrieval-Augmented Reasoning for Image-Text Interaction

2026-03-05 · Jiashu Yang, Chi Zhang, Abudukelimu Wuerkaixi, Xuxin Cheng 외 arxiv

Multimodal document question answering requires retrieving dispersed evidence from visually rich long documents and performing reliable reasoning over heterogeneous information. Existing multimodal RAG systems remain lim…

Question Answering Answer Generation Logical Reasoning