Awesome-Graph-augmented-LLM-Agent (GLA)

A curated list of awesome papers and resources on leveraging graph structures and algorithms to enhance LLM-based Agent Systems, accompanying our paper: Graph-Augmented Large Language Model Agents: Current Progress and Future Prospects.

News

• Our paper has been accepted by IEEE Intelligent Systems!

🗺️ Table of Contents

• Graph-Augmented LLM Agent System Framework
  • 🧠 Graphs for Agent Planning
  • 💾 Graphs for Agent Memory Management
  • 🛠️ Graphs for Tool Management
• Graph-Augmented LLM Multi-Agent Systems
  • 🕸️ Graphs for MAS Orchestration
  • ⚡ Graph for MAS Efficiency
  • 🛡️ Graphs for Trustworthy MAS

---

Graph-Augmented LLM Agent System Framework

🧠 Graphs for Agent Planning

Plan as a Graph

• AFlow: Automating Agentic Workflow Generation (ICLR, 2025) 📄 [paper] 💻 [code]
• VFlow: Discovering Optimal Agentic Workflows for Verilog Generation (arxiv, 2025) 📄 [paper]
• AgentKit: Structured LLM Reasoning with Dynamic Graphs (CLM, 2024) 📄 [paper] 💻 [code]
• Plan-over-Graph: Towards Parallelable LLM Agent Schedule (arxiv, 2025) 📄 [paper] 💻 [code]
• Benchmarking Agentic Workflow Generation (ICLR, 2025) 📄 [paper] 💻 [code]

Sub-task Pool as a Graph

• Can Graph Learning Improve Planning in LLM-based Agents? (NeurIPS, 2024) 📄 [paper] 💻 [code]
• Hugginggpt: Solving ai tasks with chatgpt and its friends in hugging face (NeurIPS, 2023) 📄 [paper] 💻 [code]

Reasoning Thought as a Graph

• Chain-of-thought prompting elicits reasoning in large language models (NeurIPS, 2023) 📄 [paper]
• Tree of Thoughts: Deliberate Problem Solving with Large Language Models (NeurIPS, 2023) 📄 [paper] 💻 [code]
• RATT: A Thought Structure for Coherent and Correct LLM Reasoning (AAAI, 2025) 📄 [paper]
• Graph of Thought: Solving Elaborate Problems with Large Language Models (AAAI, 2024) 📄 [paper] 💻 [code]
• Thought Graph: Generating Thought Process for Biological Reasoning (WWW, 2024) 📄 [paper] 💻 [code]
• Graph of Goal-Oriented Thoughts: Design and Implementation of LLM Agents (SOCIALIZE, 2025) 📄 [paper] 💻 [code]

Environment as a Graph

• Graphormer-Guided Task Planning: Beyond Static Rules with LLM Safety Perception (arxiv, 2025) 📄 [paper] 💻 [code]
• LocAgent: Graph-Guided LLM Agents for Code Localization (ACL, 2025) 📄 [paper] 💻 [code]

💾 Graphs for Agent Memory Management

Graph-organized Interaction Memory

• A-MEM: Agentic Memory for LLM Agents (NeurIPS, 2025) 📄 [paper] 💻 [code]
• AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents (arxiv, 2024) 📄 [paper] 💻 [code]

Graph-organized Knowledge Memory

• Synergizing LLM Agents and Knowledge Graph for Socioeconomic Prediction in LBSN (arxiv, 2024) 📄 [paper]
• KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph (ACL, 2025) 📄 [paper]

🛠️ Graphs for Tool Management

Tool Graphs for Tool Selection

• ControlLLM: Augment Language Models with Tools by Searching on Graphs (ECCV, 2024) 📄 [paper] 💻 [code]
• SciToolAgent: A Knowledge Graph-Driven Scientific Agent for Multi-Tool Integration (Nature Computational Science, 2025) 📄 [paper] 💻 [code]
• ToolNet: Connecting large language models with massive tools via tool graph (arxiv, 2024) 📄 [paper]

Tool Graphs Improve Agent Tool-use Capability

• ToolFlow: Boosting LLM Tool-Calling Through Natural and Coherent Dialogue Synthesis (NAACL, 2025) 📄 [paper]

---

Graph-Augmented LLM Multi-Agent Systems

🕸️ Graphs for MAS Orchestration

Static MAS Topology

• AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation (ICLR 2024 workshop) 📄 [paper] 💻 [code]
• Chateval: Towards better llm-based evaluators through multi-agent debate (ICLR, 2023) 📄 [paper] 💻 [code]
• Improving factuality and reasoning in language models through multiagent debate (ICML, 2024) 📄 [paper] 💻 [code]
• GPTSwarm: Language Agents as Optimizable Graphs (ICML, 2024) 📄 [paper] 💻 [code]
• Scaling Large Language Model-based Multi-Agent Collaboration (ICLR, 2025) 📄 [paper] 💻 [code]
• AFlow: Automating Agentic Workflow Generation (ICLR, 2025) 📄 [paper] 💻 [code]
• EvoFlow: Evolving Diverse Agentic Workflows On The Fly (arxiv, 2025) 📄 [paper] 💻 [code]

Task-dynamic MAS Topology

• G-Designer: Architecting Multi-agent Communication Topologies via Graph Neural Networks (ICML, 2025) 📄 [paper] 💻 [code]
• Understanding the Information Propagation Effects of Communication Topologies in LLM-based Multi-Agent Systems (EMNLP, 2025) 📄 [paper] 💻 [code]
• Assemble Your Crew: Automatic Multi-agent Communication Topology Design via Autoregressive Graph Generation (AAAI, 2026) 📄 [paper] 💻 [code]
• GNNs as Predictors of Agentic Workflow Performances (arxiv, 2025) 📄 [paper] 💻 [code]
• Multi-agent Architecture Search via Agentic Supernet (ICML, 2025) 📄 [paper] 💻 [code]
• Multi-Agent Design: Optimizing Agents with Better Prompts and Topologies (arxiv, 2025) 📄 [paper]

Process-dynamic MAS Topology

• ReSo: A Reward-driven Self-organizing LLM-based Multi-Agent System for Reasoning Tasks (arxiv, 2025) 📄 [paper] 💻 [code]
• Self-evolving Multi-agent Collaboration Networks for Software Development (ICLR, 2025) 📄 [paper] 💻 [code]
• AnyMAC: Cascading Flexible Multi-Agent Collaboration via Next-Agent Prediction (EMNLP, 2025) 📄 [paper] 💻 [code]

⚡ Graph for MAS Efficiency

Edge Redundancy in MAS

• Cut the Crap: An Economical Communication Pipeline for LLM-based Multi-Agent Systems (ICLR, 2025) 📄 [paper] 💻 [code]
• Improving Multi-Agent Debate with Sparse Communication Topology (Findings of EMNLP, 2024) 📄 [paper]
• Talk structurally, act hierarchically: A collaborative framework for llm multi-agent systems (arxiv, 2025) 📄 [paper] 💻 [code]

Node Redundancy in MAS

• AgentDropout: Dynamic Agent Elimination for Token-Efficient and High-Performance LLM-Based Multi-Agent Collaboration (ACL, 2025) 📄 [paper] 💻 [code]

Layer Redundancy in MAS

• RMoA: Optimizing Mixture-of-Agents through Diversity Maximization and Residual Compensation (Findings of ACL, 2025) 📄 [paper] 💻 [code]
• Debate Only When Necessary: Adaptive Multiagent Collaboration for Efficient LLM Reasoning (arxiv, 2025) 📄 [paper]

🛡️ Graphs for Trustworthy MAS

• G-Safeguard: A Topology-Guided Security Lens and Treatment on LLM-based Multi-agent Systems (ACL, 2025) 📄 [paper] 💻 [code]
• NetSafe: Exploring the Topological Safety of Multi-agent Network (ACL, 2025) 📄 [paper] 💻 [code]
• AgentSafe: Safeguarding Large Language Model-based Multi-agent Systems via Hierarchical Data Management (arxiv, 2025) 📄 [paper] 💻 [code]
• Goal-Aware Identification and Rectification of Misinformation in Multi-Agent Systems (arxiv, 2025) 📄 [paper] 💻 [code]
• FlowReasoner: Reinforcing Query-level Meta-agents (arxiv, 2025) 📄 [paper] 💻 [code]
• Agent-SafetyBench: Evaluating the Safety of LLM Agents (arxiv, 2024) 📄 [paper] 💻 [code]
• AgentAuditor: Human-Level Safety and Security Evaluation for LLM Agents (NeurIPS, 2025) 📄 [paper] 💻 [code]

---

Contributing

Contributions are welcome! If you have a paper or resource to add, please feel free to open a pull request.

Please ensure your submission follows this format:

• Paper Name (Conference / Journal, Year) 📄 [[paper]](link) 💻 [[code]](link)

---

Citation

If you find this repository helpful for your work, please kindly cite:

@article{liu2025graph,
  title={Graph-Augmented Large Language Model Agents: Current Progress and Future Prospects},
  author={Liu, Yixin and Zhang, Guibin and Wang, Kun and Li, Shiyuan and Pan, Shirui},
  journal={arXiv preprint arXiv:2501.02474},
  year={2025}
}