From Web Search towards Agentic Deep Research: Incentivizing Search with Reasoning Agents

2025-06-23

AI Summary

Traditional keyword search struggles with multi-step, complex information needs. To address this, we propose “Agent-Augmented Deep Research”—a novel paradigm that employs large language model–based reasoning agents to perform end-to-end deep research via autonomous task planning, iterative retrieval, knowledge integration, and dynamic feedback. We formally introduce the “test-time computational depth scaling law,” the first theoretical characterization of how computational depth in reasoning and search governs performance. Our system integrates open-source toolchains with a unified benchmarking framework and achieves significant improvements over state-of-the-art methods across multiple complex question-answering and research-oriented benchmarks (average +23.6%). This work advances information retrieval from passive pattern matching toward autonomous, agent-driven investigative reasoning, establishing both theoretical foundations and practical pathways for next-generation knowledge acquisition systems.

Technology Category

• Search and Optimization: Learning to Search
• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Information retrieval is a cornerstone of modern knowledge acquisition, enabling billions of queries each day across diverse domains. However, traditional keyword-based search engines are increasingly inadequate for handling complex, multi-step information needs. Our position is that Large Language Models (LLMs), endowed with reasoning and agentic capabilities, are ushering in a new paradigm termed Agentic Deep Research. These systems transcend conventional information search techniques by tightly integrating autonomous reasoning, iterative retrieval, and information synthesis into a dynamic feedback loop. We trace the evolution from static web search to interactive, agent-based systems that plan, explore, and learn. We also introduce a test-time scaling law to formalize the impact of computational depth on reasoning and search. Supported by benchmark results and the rise of open-source implementations, we demonstrate that Agentic Deep Research not only significantly outperforms existing approaches, but is also poised to become the dominant paradigm for future information seeking. All the related resources, including industry products, research papers, benchmark datasets, and open-source implementations, are collected for the community in https://github.com/DavidZWZ/Awesome-Deep-Research.

Problem

• Transition from keyword search to reasoning-based agentic deep research
• Address complex multi-step information needs with LLMs
• Improve information retrieval via dynamic reasoning and synthesis

Innovation

• LLMs with reasoning and agentic capabilities
• Autonomous reasoning and iterative retrieval integration
• Test-time scaling law for computational depth impact