Research Paper (preprint): “Deep Research Agents: A Systematic Examination and Roadmap”

The research paper (shared below) was recently posted on arXiv

Title

Deep Research Agents: A Systematic Examination and Roadmap

Authors

Yuxuan Huang
University of Liverpool

Yihang Chen
Huawei Noah’s Ark Lab

Haozheng Zhang
Huawei Noah’s Ark Lab

Kang Li
University of Oxford

Meng Fang
University of Liverpool

Linyi Yang
University College London

Xiaoguang Li
University of Liverpool

Lifeng Shang
University of Liverpool

Songcen Xu
Huawei Noah’s Ark Lab

Jianye Hao
Huawei Noah’s Ark Lab

Kun Shao
Huawei Noah’s Ark Lab

Jun Wang
University College London

Source

via arXiv

DOI: 10.48550/arXiv.2506.18096

Abstract

The rapid progress of Large Language Models (LLMs) has given rise to a new category of autonomous AI systems, referred to as Deep Research (DR) agents. These agents are designed to tackle complex, multi-turn informational research tasks by leveraging a combination of dynamic reasoning, adaptive long-horizon planning, multi-hop information retrieval, iterative tool use, and the generation of structured analytical reports. In this paper, we conduct a detailed analysis of the foundational technologies and architectural components that constitute Deep Research agents. We begin by reviewing information acquisition strategies, contrasting API-based retrieval methods with browser-based exploration. We then examine modular tool-use frameworks, including code execution, multimodal input processing, and the integration of Model Context Protocols (MCPs) to support extensibility and ecosystem development. To systematize existing approaches, we propose a taxonomy that differentiates between static and dynamic workflows, and we classify agent architectures based on planning strategies and agent composition, including single-agent and multi-agent configurations. We also provide a critical evaluation of current benchmarks, highlighting key limitations such as restricted access to external knowledge, sequential execution inefficiencies, and misalignment between evaluation metrics and the practical objectives of DR agents. Finally, we outline open challenges and promising directions for future research. A curated and continuously updated repository of DR agent research is available.

Direct to Full Text Article
26 pages; PDF.

Filed under: Journal Articles, News, Open Access, Reports