CS-Agent: LLM-based Community Search via Dual-agent Collaboration

2025-08-13

AI Summary

Large language models (LLMs) exhibit significant output bias and unreliable performance in graph community search tasks. Method: This paper introduces GraphCS, the first LLM-oriented community search benchmark, and proposes a Solver-Validator dual-agent collaborative framework: a prompt-engineered Solver performs zero-shot community identification; a dynamic Validator assesses outputs and provides feedback; and a Decider module aggregates multi-round reasoning results. The approach requires no fine-tuning, integrating iterative refinement with structured decision-making. Contribution/Results: Experiments on multiple real-world graph datasets demonstrate substantial improvements in accuracy and robustness. This work constitutes the first systematic validation of LLMs’ feasibility and effectiveness for graph-structured analysis—without parameter adjustment—thereby establishing a novel paradigm for LLM-augmented graph learning.

Technology Category

• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Search and Optimization: Learning to Search
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information

Abstract

Large Language Models (LLMs) have demonstrated remarkable capabilities in natural language processing tasks, yet their application to graph structure analysis, particularly in community search, remains underexplored. Community search, a fundamental task in graph analysis, aims to identify groups of nodes with dense interconnections, which is crucial for understanding the macroscopic structure of graphs. In this paper, we propose GraphCS, a comprehensive benchmark designed to evaluate the performance of LLMs in community search tasks. Our experiments reveal that while LLMs exhibit preliminary potential, they frequently fail to return meaningful results and suffer from output bias. To address these limitations, we introduce CS-Agent, a dual-agent collaborative framework to enhance LLM-based community search. CS-Agent leverages the complementary strengths of two LLMs acting as Solver and Validator. Through iterative feedback and refinement, CS-Agent dynamically refines initial results without fine-tuning or additional training. After the multi-round dialogue, Decider module selects the optimal community. Extensive experiments demonstrate that CS-Agent significantly improves the quality and stability of identified communities compared to baseline methods. To our knowledge, this is the first work to apply LLMs to community search, bridging the gap between LLMs and graph analysis while providing a robust and adaptive solution for real-world applications.

Problem

• Enhancing LLM-based community search in graph analysis
• Addressing output bias and low-quality results in LLMs
• Developing dual-agent collaboration for dynamic community refinement

Innovation

• Dual-agent collaboration enhances community search
• Iterative feedback refines results without training
• Decider module selects optimal community dynamically