FlowSearch: Advancing deep research with dynamic structured knowledge flow

2025-10-09

AI Summary

Deep research faces challenges including multi-step reasoning complexity and difficulty in cross-domain knowledge integration. To address these, we propose a multi-agent collaborative framework that models knowledge flow via a dynamic graph structure, enabling hierarchical subtask-driven decomposition, adaptive planning, and parallel exploration. A real-time feedback mechanism supports online evolution and regulation of knowledge flow. Our core contribution lies in the tight integration of dynamically constructed structured knowledge flow with multi-agent collaborative reasoning—balancing broad exploration and deep deductive reasoning. Evaluated on diverse benchmarks—including GAIA, HLE, GPQA, and TRQA—our approach achieves state-of-the-art performance across both general and scientific domains, significantly improving efficiency and accuracy in interdisciplinary research.

Technology Category

• Knowledge Representation and Reasoning: Computational Complexity of Reasoning
• Multiagent Systems: Multiagent Planning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs

Abstract

Deep research is an inherently challenging task that demands both breadth and depth of thinking. It involves navigating diverse knowledge spaces and reasoning over complex, multi-step dependencies, which presents substantial challenges for agentic systems. To address this, we propose FlowSearch, a multi-agent framework that actively constructs and evolves a dynamic structured knowledge flow to drive subtask execution and reasoning. FlowSearch is capable of strategically planning and expanding the knowledge flow to enable parallel exploration and hierarchical task decomposition, while also adjusting the knowledge flow in real time based on feedback from intermediate reasoning outcomes and insights. FlowSearch achieves state-of-the-art performance on both general and scientific benchmarks, including GAIA, HLE, GPQA and TRQA, demonstrating its effectiveness in multi-disciplinary research scenarios and its potential to advance scientific discovery. The code is available at https://github.com/Alpha-Innovator/InternAgent.

Problem

• Enables parallel exploration and hierarchical task decomposition
• Adjusts knowledge flow dynamically using reasoning feedback
• Addresses multi-disciplinary research with structured knowledge evolution

Innovation

• Multi-agent framework constructs dynamic structured knowledge flow
• Strategically plans knowledge flow for parallel exploration
• Real-time adjustment based on intermediate reasoning feedback