arXiv Computation and Language November 4, 2026 By Tiansheng Hu, Yichuan Wang, James V. Talwar, Nicholas Tomlin, Chen Zhao, Sewon Min

DR Pixel: Page-Native Deep Research with Vision-Language Models

Read the original on arXiv Computation and Language →

DR Pixel introduces a page‑native deep research agent that uses rendered document page images as the sole unit for retrieval and reasoning, bypassing OCR. The open‑source system operates on 1.06 M pages and, without training, outperforms OCR‑based baselines on GPT‑5.4, Gemini‑3.5‑Flash, and Qwen3.6‑27B while requiring fewer tool calls. Training with counterfactual credit reweighting further boosts retrieval recall by ~15 points and accuracy by >10 points on the DRPixelBench benchmark, and brings the model within 1.8 points of an oracle on the ViDoRe‑V3 out‑of‑distribution test.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Computation and Language.

arXiv Computation and Language

Aug 21

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

arXiv:2604. 13731v2 Announce Type: replace Abstract: Multi-page Document Visual Question Answering requires reasoning over semantics, layouts, and visual elements in long, visually dense documents.

By Yuanlei Zheng, Pei Fu, Hang Li, Ziyang Wang, Yuyi Zhang, Wenyu Ruan, Xiaojin Zhang, Zhongyu Wei, Zhenbo Luo, Jian Luan, Wei Chen, Xiang Bai

rag agents computer-vision nlp benchmarks

arXiv AI

Oct 16

DocPilot: Interactive Agent Orchestration and Training for Long Visual Document Understanding

DocPilot is an agentic framework for long visual document understanding that orchestrates evidence seeking through page-level navigation and fine-grained localization, integrating visual inspection, code-based numerical reasoning, and cross-modal reasoning. It employs curriculum-based agentic training to enable small language models to learn complex document solving, progressing from single-ability learning to adaptive orchestration. Experiments on benchmarks such as MMLongBench-Doc, DocBench, FinRAGBench-V, and MADQA show that DocPilot achieves strong gains with frontier VLM backbones, and its 4B variant competes with much larger models, demonstrating effectiveness and generalization across diverse long-document reasoning tasks.

By Zhuowen Liang, Xiaotian Lin, Yuyu Luo, Haixun Wang, Nan Tang

llms agents nlp multimodal

arXiv Computation and Language

Sep 1

OCR-MetaReasoning Benchmark: Evaluating the Meta-Reasoning Ability of MLLMs in Text-Rich Image Understanding

arXiv:2608.30678v1 Announce Type: new Abstract: Text-rich image understanding requires multimodal large language models (MLLMs) to organize OCR (Optical Character Recognition)-grounded evidence acros...

By Gengxu Li, Yuan Wu, Yi Chang

llms computer-vision multimodal benchmarks

arXiv AI

Sep 7

SciDocBench: A Workflow-Centered Benchmark and Data Pipeline for Scientific Document Understanding

SciDocBench is a workflow-centered benchmark for scientific document understanding that includes 124 expert-authored questions across seven capability groups and 19 subtasks in five scientific domains. Each question is evaluated under four conditions—English or Chinese, all-images-first or interleaved document representations—resulting in 496 evaluation instances. The benchmark is paired with SciDocIR, a typed evidence-graph representation, and SciDocDataset, a collection of 15K fine-tuning and 8K reinforcement-learning samples, forming an evaluation-to-training framework for scientific-document assistants.

By Shenxi Wu, Yuhong Liu, Haosong Zhang, Tongjin Zou, Yanxun Zhang, Gaochang Chen, Dun Liang, Jiaqi Wang, Zhecan James Wang, Yuhang Zang, Dahua Lin

fine-tuning multimodal benchmarks data-engineering

arXiv Machine Learning

Aug 12

InSight-doc: Agentic Visual Perception for Long-Document Understanding

arXiv:2608. 10628v1 Announce Type: cross Abstract: Long-document understanding often requires reasoning over many visually rich pages, making inference costly and prone to context rot.

By Kaican Li, Weiyan Xie, Lewei Yao, Jiannan Wu, Lanqing Hong, Yongxiang Huang, Nevin L. Zhang

agents reinforcement-learning fine-tuning efficiency benchmarks safety

arXiv Computer Vision

Oct 27

OvisOCR3 Technical Report

arXiv:2610.33066v1 Announce Type: new Abstract: We introduce OvisOCR3, a 0.8B model for a broad range of OCR-related tasks. OvisOCR3 unifies document parsing, general OCR, key information extraction...

By Yu Xia, Yinglun Li, Yuhui Chen, Jinlong Huang, Yuyang Wanyan, Shengze Shi, Yifei Song, Yue Duan, Shubo Lin, Jianshan Zhao, En Lin, Shiyin Lu, Qing-Guo Chen, Zhao Xu, Weihua Luo

llms computer-vision reinforcement-learning efficiency multimodal benchmarks