Summary

This repository provides a systematic and comprehensive survey of "Deep Research" (DR), a rapidly evolving field. It aims to serve as a foundational resource for researchers and practitioners by organizing key papers across critical components, training paradigms, and benchmarks. The project offers a structured overview, a roadmap for understanding DR, and insights into future directions, with a commitment to continuous updates.

How It Works

The project functions as a curated collection of academic papers related to Deep Research. It systematically categorizes research into key areas such as Query Planning, Information Acquisition, Knowledge Boundary Retrieval, Timing, Agentic End-to-End Reinforcement Learning, Supervised Fine-tuning, and Datasets & Benchmarks. This organization facilitates a deep dive into the foundational concepts, practical techniques, challenges, and emerging trends within DR. The core value lies in its comprehensive cataloging and ongoing maintenance to reflect the field's dynamic progress.

Quick Start & Requirements

This repository is a survey and reading list, not a software project. It does not have installation or execution requirements. Key resources include:

• Survey PDF: https://github.com/mangopy/Deep-Research-Survey/blob/main/Deep-Research-Survey.pdf
• Web Version: https://deep-research-survey.github.io/
• Preprint: https://arxiv.org/abs/2512.02038

Highlighted Details

• Comprehensive categorization of Deep Research papers across eight key areas, including agentic RL, information acquisition, and benchmarks.
• Extensive reading lists provide paper titles, venues, dates, and direct URLs for hundreds of relevant research articles.
• Includes links to the full survey PDF, a dedicated website, and preprints on platforms like arXiv and Preprints.org.
• The project is actively maintained with a commitment to continuous updates reflecting the latest advancements in Deep Research.

Maintenance & Community

The repository is marked as "Active" and commits to continuous updates. Contact information is provided via email (zhengliang.shii@gmail.com, shizhl@mail.sdu.edu.cn) for inquiries and discussions. The project encourages GitHub stars for staying updated.

Licensing & Compatibility

No specific open-source license is mentioned in the provided README content. Compatibility for commercial use or closed-source linking is therefore undetermined.

Limitations & Caveats

This repository serves as a curated academic resource rather than a runnable software tool. Some paper dates listed are in the future (e.g., 2025, 2026), indicating a forward-looking scope or potential placeholders. The "Reading List will be updated as soon as possible!" note suggests that certain sections may still be under development or incomplete.