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

• URL: http://arxiv.org/abs/2606.26122v1
• Date: Wed, 27 May 2026 21:21:42 GMT
• Title: DocArena: Turning Raw Documents into Controllable Training Environments for Document Search Agents
• Abstract summary: DocArena transforms raw document collections into training environments for search agents without any human annotation.<n>We introduce DocArena-79K with QA pairs from 8,336 documents spanning 16 domains and 49 languages.<n>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.
• Score: 44.01307558575661
• License: http://creativecommons.org/licenses/by-nc-sa/4.0/
• 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.

Related papers

• Training Documents Reranker with Search Rubrics for Deep Research Agent [54.08506174105984]
  We propose search-oriented rubrics that textitexplicitly define the requirements that high-quality document sets should satisfy.<n>We further train a document reranker textbfRubricRanker to select a high-quality subset from retrieved documents.
  arXiv  Detail & Related papers  (2026-08-04T12:11:14Z)
• GrepSeek: Training Search Agents for Direct Corpus Interaction [66.69568141699311]
  GrepSeek is an optimized direct corpus interaction (DCI) search agent that trains a compact search agent to find, filter, and compose evidence from large text corpora.<n>To make DCI practical at scale, we further use a semantics-preserving sharded-parallel execution engine that accelerates shell-based retrieval by up to $7.6times$ while preserving byte-exact equivalence with sequential execution of the shell command.
  arXiv  Detail & Related papers  (2026-05-28T03:37:33Z)
• Navigating Large-Scale Document Collections: MuDABench for Multi-Document Analytical QA [25.155696504567718]
  This paper introduces the task of analytical question answering over large, semi-structured document collections.<n>We present MuDABench, a benchmark for multi-document analytical QA.
  arXiv  Detail & Related papers  (2026-04-24T05:28:51Z)
• DocSeeker: Structured Visual Reasoning with Evidence Grounding for Long Document Understanding [63.257540233507626]
  We propose a paradigm that requires the model to execute a structured Analysis, Localization and Reasoning workflow.<n>We show it robustly generalizes from short-page training to ultra-long documents and is naturally synergistic with visual Retrieval-Augmented Generation systems.
  arXiv  Detail & Related papers  (2026-04-14T14:39:26Z)
• Model-Document Protocol for AI Search [11.377241012645994]
  We introduce the Model-Document Protocol (MDP), a general framework that formalizes how raw text is bridged to large language models (LLMs)<n>Rather than treating retrieval as passage fetching, MDP defines multiple pathways that transform unstructured documents into task-specific, LLM-ready inputs.<n>As an instantiation, we present MDP-Agent, which realizes the protocol through an agentic process.
  arXiv  Detail & Related papers  (2025-10-29T04:29:17Z)
• Doc-Researcher: A Unified System for Multimodal Document Parsing and Deep Research [31.973886754355547]
  Doc-Researcher is a unified system that bridges the gap between text-only, vision-only, and hybrid paradigms.<n>We introduce M4DocBench, the first benchmark for Multi-modal, Multi-hop, Multi-document, and Multi-turn deep research.<n>Experiments demonstrate Doc-Researcher achieves 50.6% accuracy, 3.4xbetter than state-of-the-art baselines.
  arXiv  Detail & Related papers  (2025-10-24T16:07:54Z)
• MMESGBench: Pioneering Multimodal Understanding and Complex Reasoning Benchmark for ESG Tasks [56.350173737493215]
  Environmental, Social, and Governance (ESG) reports are essential for evaluating sustainability practices, ensuring regulatory compliance, and promoting financial transparency.<n>MMESGBench is a first-of-its-kind benchmark dataset to evaluate multimodal understanding and complex reasoning across structurally diverse and multi-source ESG documents.<n>MMESGBench comprises 933 validated QA pairs derived from 45 ESG documents, spanning across seven distinct document types and three major ESG source categories.
  arXiv  Detail & Related papers  (2025-07-25T03:58:07Z)
• Beyond Retrieval: Joint Supervision and Multimodal Document Ranking for Textbook Question Answering [3.6799953119508735]
  We propose a novel approach to multimodal textbook question answering by introducing a mechanism for enhancing semantic representations.<n>Our model, Joint Embedding Training With Ranking Supervision for Textbook Question Answering (JETRTQA), is a multimodal learning framework built on a retriever-generator architecture.<n>We evaluate our method on the CK12-QA dataset and demonstrate that it significantly improves the discrimination between informative and irrelevant documents.
  arXiv  Detail & Related papers  (2025-05-17T13:23:54Z)
• Unified Pretraining Framework for Document Understanding [52.224359498792836]
  We present UDoc, a new unified pretraining framework for document understanding. UDoc is designed to support most document understanding tasks, extending the Transformer to take multimodal embeddings as input. An important feature of UDoc is that it learns a generic representation by making use of three self-supervised losses.
  arXiv  Detail & Related papers  (2022-04-22T21:47:04Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.