Technology Category
Multiagent Systems: Modeling other AgentsCognitive Modeling & Cognitive Systems: Agent ArchitecturesHumans and AI: Human-Aware Planning and Behavior Prediction
Graphs Meet AI Agents: Taxonomy, Progress, and Future Opportunities
🤖 AI Summary
AI agents face challenges in complex real-world tasks, including brittle planning, unreliable memory, and inefficient multi-agent coordination. Method: This paper proposes the Graph-Augmented Agents paradigm, systematically integrating graph neural networks, knowledge graphs, reinforcement learning, and large language models to uniformly represent task logic, memory trajectories, and collaboration relationships via graph-structured modeling. Contribution/Results: We introduce the first systematic taxonomy unifying graphs and agents, revealing graphs’ structural advantages in relational abstraction, dynamic reasoning, and interpretable coordination. Empirical evaluation and comprehensive survey demonstrate substantial improvements in long-horizon planning consistency, memory retrieval accuracy, and multi-agent coordination efficiency. To foster community advancement, we establish an open-source survey platform that continuously curates state-of-the-art graph-agent works and benchmark resources, providing both theoretical foundations and practical guidelines for this interdisciplinary field.
Technology Category
Application Category
📝 Abstract
AI agents have experienced a paradigm shift, from early dominance by reinforcement learning (RL) to the rise of agents powered by large language models (LLMs), and now further advancing towards a synergistic fusion of RL and LLM capabilities. This progression has endowed AI agents with increasingly strong abilities. Despite these advances, to accomplish complex real-world tasks, agents are required to plan and execute effectively, maintain reliable memory, and coordinate smoothly with other agents. Achieving these capabilities involves contending with ever-present intricate information, operations, and interactions. In light of this challenge, data structurization can play a promising role by transforming intricate and disorganized data into well-structured forms that agents can more effectively understand and process. In this context, graphs, with their natural advantage in organizing, managing, and harnessing intricate data relationships, present a powerful data paradigm for structurization to support the capabilities demanded by advanced AI agents. To this end, this survey presents a first systematic review of how graphs can empower AI agents. Specifically, we explore the integration of graph techniques with core agent functionalities, highlight notable applications, and identify prospective avenues for future research. By comprehensively surveying this burgeoning intersection, we hope to inspire the development of next-generation AI agents equipped to tackle increasingly sophisticated challenges with graphs. Related resources are collected and continuously updated for the community in the Github link.
Problem
Research questions and friction points this paper is trying to address.
Enhancing AI agent capabilities with graph-structured data
Integrating graphs for improved planning and memory in agents
Exploring graph techniques to support multi-agent coordination
Innovation
Methods, ideas, or system contributions that make the work stand out.
Fusion of RL and LLM for AI agents
Graphs structure data for agent processing
Systematic review of graph-powered AI agents
🔎 Similar Papers
