A Survey of LLM-based Deep Search Agents: Paradigm, Optimization, Evaluation, and Challenges

2025-08-03

AI Summary

Existing research on large language model (LLM)-based deep search agents lacks a unified taxonomy and standardized evaluation framework, hindering systematic progress and cross-study comparison. Method: This work introduces the first comprehensive analytical framework covering architectural design, optimization mechanisms—including dynamic planning, multi-turn reasoning, and autonomous decision-making—application scenarios, and evaluation methodologies. It identifies critical open challenges: contextual modeling, long-horizon consistency, interpretability, and the absence of robust evaluation benchmarks. Contribution/Results: The study proposes a forward-looking research agenda centered on autonomous, multi-turn dynamic information retrieval. To support reproducible and scalable research, it publicly releases an annotated survey repository and a hierarchical classification toolkit. This constitutes the first structured, domain-specific survey and provides foundational, reusable research infrastructure for the LLM-based deep search agent community.

Technology Category

• Search and Optimization: Learning to Search
• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

The advent of Large Language Models (LLMs) has significantly revolutionized web search. The emergence of LLM-based Search Agents marks a pivotal shift towards deeper, dynamic, autonomous information seeking. These agents can comprehend user intentions and environmental context and execute multi-turn retrieval with dynamic planning, extending search capabilities far beyond the web. Leading examples like OpenAI's Deep Research highlight their potential for deep information mining and real-world applications. This survey provides the first systematic analysis of search agents. We comprehensively analyze and categorize existing works from the perspectives of architecture, optimization, application, and evaluation, ultimately identifying critical open challenges and outlining promising future research directions in this rapidly evolving field. Our repository is available on https://github.com/YunjiaXi/Awesome-Search-Agent-Papers.

Problem

• Surveying LLM-based search agents' paradigms and challenges
• Analyzing optimization and evaluation of deep search agents
• Identifying future research directions for autonomous search

Innovation

• LLM-based dynamic planning for multi-turn retrieval
• Comprehensive analysis of search agent architectures
• Systematic evaluation of deep search applications