2026-06-26 04:00 UTCOriginal source2 min readUpdated: 2026-06-26 08:05 UTC

DocArena: Turning Raw Documents into Controllable Training Environments for Document Search Agents

DocArena is a fully automated data curation pipeline that uses multimodal large language models (MLLMs) to transform raw documents into controllable, scalable training environments for document search agents. It requires no human annotation, generates reasoning-intensive QA pairs, and produces the DocArena-79K dataset spanning 8,336 documents across 16 domains and 49 languages. Experiments show that agents trained on DocArena achieve state-of-the-art performance on both retrieval accuracy and QA quality.

SourcearXiv Computer VisionAuthor: Jiamian Wang, Ruiyi Zhang, Tong Yu, Jing Shi, Samyadeep Basu, Rajiv Jain, Zhiqiang Tao, Tong Sun

Article intelligence

EngineersAdvanced

Key points

DocArena automates the construction of training environments from raw documents using MLLM-based visual perception without human annotation.
The DocArena-79K dataset covers 16 domains and 49 languages with 8,336 documents.
The Doc-Search agent architecture decouples visual perception from the policy model, enabling text-based LLMs for multimodal retrieval.
Agents trained on DocArena outperform prior methods on multiple benchmarks, demonstrating controllability and effectiveness.

Why it matters

This matters because docArena automates the construction of training environments from raw documents using MLLM-based visual perception without human annotation.

Technical impact

May affect model selection, inference cost, product capability, and evaluation benchmarks.

This panel is AI-generated and reviewed for accuracy.

[2606.26122] DocArena: Turning Raw Documents into Controllable Training Environments for Document Search Agents

[Submitted on 27 May 2026]

Title:DocArena: Turning Raw Documents into Controllable Training Environments for Document Search Agents

View a PDF of the paper titled DocArena: Turning Raw Documents into Controllable Training Environments for Document Search Agents, by Jiamian Wang and 7 other authors

View PDF HTML (experimental)

Abstract:Recent methods train search agents via reinforcement learning from (question, answer, evidence) tuples without requiring expert trajectories. The tuples serve as the training environment, and whose properties directly shape what search strategies and generalization abilities the agent can develop. While prior works have made encouraging progress in improving training data quality, existing environments remain predominantly text-based and existing approaches can struggle to construct training environments that are controllable, scalable, and account for multimodal data. Given this, we propose DocArena, a fully automated data curation pipeline building on the practical need for multimodal document search and question-answering. It transforms raw document collections into training environments for search agents without any human annotation. The pipeline first structures and indexes documents through MLLM-based visual perception, then profiles and leverage the cross-page information distribution to construct reasoning-intensive QA pairs, as well as performs cascaded quality assurance operations via MLLM. We introduce DocArena-79K with QA pairs from 8,336 documents spanning 16 domains and 49 languages. We further design a Doc-Search agent infrastructure that decouples visual perception from the policy model, allowing text-based LLMs to serve as the reasoning backbone for multimodal document retrieval and QA. Under a unified evaluation framework where only the policy model differs, experiments on six multimodal document scenarios and seven text-based QA benchmarks show that agents trained on DocArena data achieve the best performance on both retrieval accuracy and QA quality. Further analysis on agent search behaviors confirms the effectiveness and controllability of the constructed training environment.

Comments: search agent for documents

Subjects:

Computer Vision and Pattern Recognition (cs.CV)

Cite as: arXiv:2606.26122 [cs.CV]

(or arXiv:2606.26122v1 [cs.CV] for this version)

https://doi.org/10.48550/arXiv.2606.26122

arXiv-issued DOI via DataCite

Submission history

From: Samyadeep Basu [view email] [v1] Wed, 27 May 2026 21:21:42 UTC (1,191 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled DocArena: Turning Raw Documents into Controllable Training Environments for Document Search Agents, by Jiamian Wang and 7 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.CV

new | recent | 2026-06

Change to browse by:

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)