Agent¶

GitHub 资源 ¶

• Awesome SDKs for AI Agents - AI 代理 SDK 的精选列表
• GitHub - kyrolabs/awesome-agents: 🤖 Awesome list of AI Agents

开源 AI 代理项目 ¶

common agent¶

1. GitHub - camel-ai/camel: 🐫 CAMEL: Finding the Scaling Law of Agents. The first and the best multi-agent framework. https://www.camel-ai.org
2. GitHub - myshell-ai/AIlice: AIlice is a fully autonomous, general-purpose AI agent.
3. GitHub - microsoft/autogen: A programming framework for agentic AI 🤖 PyPi: autogen-agentchat Discord: https://aka.ms/autogen-discord Office Hour: https://aka.ms/autogen-officehour
4. GitHub - jbexta/AgentPilot: A versatile workflow automation platform to create, organize, and execute AI workflows, from a single LLM to complex AI-driven workflows.
5. GitHub - DataBassGit/AgentForge: Extensible AGI Framework
6. GitHub - reworkd/AgentGPT: 🤖 Assemble, configure, and deploy autonomous AI Agents in your browser.
7. GitHub - satellitecomponent/Neurite: Fractal Graph-of-Thought. Rhizomatic Mind-Mapping for Ai-Agents, Web-Links, Notes, and Code.
8. GitHub - geekan/MetaGPT: 🌟 The Multi-Agent Framework: First AI Software Company, Towards Natural Language Programming

others¶

1. GitHub - calcom/cal.com: Scheduling infrastructure for absolutely everyone.
2. GitHub - HumanSignal/Adala: Adala: Autonomous DAta (Labeling) Agent framework

role play¶

1. GitHub - crewAIInc/crewAI: Framework for orchestrating role-playing, autonomous AI agents. By fostering collaborative intelligence, CrewAI empowers agents to work together seamlessly, tackling complex tasks.

code¶

1. GitHub - sourcegraph/cody: Type less, code more: Cody is an AI code assistant that uses advanced search and codebase context to help you write and fix code.
2. GitHub - ajhous44/cody: Cody the coding ai assistant
3. GitHub - continuedev/continue: ⏩ Create, share, and use custom AI code assistants with our open-source IDE extensions and hub of models, rules, prompts, docs, and other building blocks
4. GitHub - Aider-AI/aider: aider is AI pair programming in your terminal

不更新的 ¶

1. GitHub - codefuse-ai/codefuse-chatbot: An intelligent assistant serving the entire software development lifecycle, powered by a Multi-Agent Framework, working with DevOps Toolkits, Code&Doc Repo RAG, etc.
2. GitHub - ennucore/clippinator: AI programming assistant
3. GitHub - ur-whitelab/chemcrow-public: Chemcrow
4. GitHub - Technion-Kishony-lab/data-to-paper: data-to-paper: Backward-traceable AI-driven scientific research
5. GitHub - stitionai/devika: Devika is an Agentic AI Software Engineer that can understand high-level human instructions, break them down into steps, research relevant information, and write code to achieve the given objective. Devika aims to be a competitive open-source alternative to Devin by Cognition AI. [⚠️ DEVIKA DOES NOT HAVE AN OFFICIAL WEBSITE ⚠️]
6. GitHub - jina-ai/dev-gpt: Your Virtual Development Team
7. GitHub - melih-unsal/DemoGPT: 🤖 Everything you need to create an LLM Agent—tools, prompts, frameworks, and models—all in one place.
8. GitHub - Farama-Foundation/chatarena: ChatArena (or Chat Arena) is a Multi-Agent Language Game Environments for LLMs. The goal is to develop communication and collaboration capabilities of AIs.
9. GitHub - OpenBMB/ChatDev: Create Customized Software using Natural Language Idea (through LLM-powered Multi-Agent Collaboration)
10. GitHub - seahyinghang8/blinky: An open-source debugging agent in VSCode
11. GitHub - BloopAI/bloop: bloop is a fast code search engine written in Rust.
12. GitHub - krohling/bondai
13. GitHub - xeol-io/bumpgen: bumpgen is an AI agent that upgrades npm packages
14. GitHub - yoheinakajima/babyagi
15. GitHub - pgalko/BambooAI: A Python library powered by Language Models (LLMs) for conversational data discovery and analysis.
16. GitHub - AutoPackAI/beebot: An Autonomous AI Agent that works
17. GitHub - stepanogil/autonomous-hr-chatbot: An autonomous HR agent that can answer user queries using tools
18. GitHub - irgolic/AutoPR: Run AI-powered workflows over your codebase
19. GitHub - emrgnt-cmplxty/automata: Automata: A self-coding agent
20. GitHub - aiwaves-cn/agents: An Open-source Framework for Data-centric, Self-evolving Autonomous Language Agents
21. GitHub - eumemic/ai-legion: An LLM-powered autonomous agent platform
22. GitHub - LehengTHU/Agent4Rec: [SIGIR 2024 perspective] The implementation of paper "On Generative Agents in Recommendation"

Paper 汇总 ¶

1. Bergman, S., Ji, Z., Kermarrec, A.-M., Petrescu, D., Pires, R., Randl, M., & Vos, M. de. (2025). Leveraging Approximate Caching for Faster Retrieval-Augmented Generation. https://doi.org/10.1145/3721146.3721941
2. Cai, Y., Guo, Z., Pei, Y., Bian, W., & Zheng, W. (2024). SimGRAG: Leveraging Similar Subgraphs for Knowledge Graphs Driven Retrieval-Augmented Generation (No. arXiv: 2412.15272). arXiv. https://doi.org/10.48550/arXiv.2412.15272
3. Chen, B., Guo, Z., Yang, Z., Chen, Y., Chen, J., Liu, Z., Shi, C., & Yang, C. (2025). PathRAG: Pruning Graph-based Retrieval Augmented Generation with Relational Paths (No. arXiv: 2502.14902). arXiv. https://doi.org/10.48550/arXiv.2502.14902
4. Cheng, Y., Zhao, Y., Zhu, J., Liu, Y., Sun, X., & Li, X. (2025). Human Cognition Inspired RAG with Knowledge Graph for Complex Problem Solving (No. arXiv: 2503.06567). arXiv. https://doi.org/10.48550/arXiv.2503.06567
5. Geng, X., Wang, H., Wang, J., Liu, W., & Li, R. (2025). Enhancing RAG with Active Learning on Conversation Records: Reject Incapables and Answer Capables (No. arXiv: 2502.09073). arXiv. https://doi.org/10.48550/arXiv.2502.09073
6. Guan, X., Zeng, J., Meng, F., Xin, C., Lu, Y., Lin, H., Han, X., Sun, L., & Zhou, J. (2025). DeepRAG: Thinking to Retrieval Step by Step for Large Language Models (No. arXiv: 2502.01142). arXiv. https://doi.org/10.48550/arXiv.2502.01142
7. Gutiérrez, B. J., Shu, Y., Qi, W., Zhou, S., & Su, Y. (2025). From RAG to Memory: Non-Parametric Continual Learning for Large Language Models (No. arXiv: 2502.14802). arXiv. https://doi.org/10.48550/arXiv.2502.14802
8. Han, H., Shomer, H., Wang, Y., Lei, Y., Guo, K., Hua, Z., Long, B., Liu, H., & Tang, J. (2025). RAG vs. GraphRAG: A Systematic Evaluation and Key Insights (No. arXiv: 2502.11371). arXiv. https://doi.org/10.48550/arXiv.2502.11371
9. Han, H., Wang, Y., Shomer, H., Guo, K., Ding, J., Lei, Y., Halappanavar, M., Rossi, R. A., Mukherjee, S., Tang, X., He, Q., Hua, Z., Long, B., Zhao, T., Shah, N., Javari, A., Xia, Y., & Tang, J. (2025). Retrieval-Augmented Generation with Graphs (GraphRAG) (No. arXiv: 2501.00309). arXiv. https://doi.org/10.48550/arXiv.2501.00309
10. Huang, H., Huang, Y., Yang, J., Pan, Z., Chen, Y., Ma, K., Chen, H., & Cheng, J. (2025). Retrieval-Augmented Generation with Hierarchical Knowledge (No. arXiv: 2503.10150). arXiv. https://doi.org/10.48550/arXiv.2503.10150
11. Huang, S., Ma, Z., Du, J., Meng, C., Wang, W., Leng, J., Guo, M., & Lin, Z. (2025). Gumbel Reranking: Differentiable End-to-End Reranker Optimization (No. arXiv: 2502.11116). arXiv. https://doi.org/10.48550/arXiv.2502.11116
12. Huang, Y., Zhang, S., & Xiao, X. (2025). KET-RAG: A Cost-Efficient Multi-Granular Indexing Framework for Graph-RAG (No. arXiv: 2502.09304). arXiv. https://doi.org/10.48550/arXiv.2502.09304
13. Lee, M.-C., Zhu, Q., Mavromatis, C., Han, Z., Adeshina, S., Ioannidis, V. N., Rangwala, H., & Faloutsos, C. (2024). HybGRAG: Hybrid Retrieval-Augmented Generation on Textual and Relational Knowledge Bases (No. arXiv: 2412.16311). arXiv. https://doi.org/10.48550/arXiv.2412.16311
14. Li, M., Gaussier, E., & Zhou, G. (2025). Enhanced Retrieval of Long Documents: Leveraging Fine-Grained Block Representations with Large Language Models (No. arXiv: 2501.17039). arXiv. https://doi.org/10.48550/arXiv.2501.17039
15. Li, X., Cao, Y., Ma, Y., & Sun, A. (2024). Long Context vs. RAG for LLMs: An Evaluation and Revisits (No. arXiv: 2501.01880). arXiv. https://doi.org/10.48550/arXiv.2501.01880
16. Lin, C.-Y., Kamahori, K., Liu, Y., Shi, X., Kashyap, M., Gu, Y., Shao, R., Ye, Z., Zhu, K., Wang, S., Krishnamurthy, A., Kadekodi, R., Ceze, L., & Kasikci, B. (2025). TeleRAG: Efficient Retrieval-Augmented Generation Inference with Lookahead Retrieval (No. arXiv: 2502.20969). arXiv. https://doi.org/10.48550/arXiv.2502.20969
17. Liu, H., Wang, Z., Chen, X., Li, Z., Xiong, F., Yu, Q., & Zhang, W. (2025). HopRAG: Multi-Hop Reasoning for Logic-Aware Retrieval-Augmented Generation (No. arXiv: 2502.12442). arXiv. https://doi.org/10.48550/arXiv.2502.12442
18. Lumer, E., Basavaraju, P. H., Mason, M., Burke, J. A., & Subbiah, V. K. (2025). Graph RAG-Tool Fusion (No. arXiv: 2502.07223). arXiv. https://doi.org/10.48550/arXiv.2502.07223
19. Luo, L., Zhao, Z., Haffari, G., Phung, D., Gong, C., & Pan, S. (2025). GFM-RAG: Graph Foundation Model for Retrieval Augmented Generation (No. arXiv: 2502.01113). arXiv. https://doi.org/10.48550/arXiv.2502.01113
20. Mahalingam, A., Gande, V. K., Chadha, A., Jain, V., & Chaudhary, D. (2024). SKETCH: Structured Knowledge Enhanced Text Comprehension for Holistic Retrieval (No. arXiv: 2412.15443). arXiv. https://doi.org/10.48550/arXiv.2412.15443
21. Mukherjee, M., Kim, S., Chen, X., Luo, D., Yu, T., & Mai, T. (2025). From Documents to Dialogue: Building KG-RAG Enhanced AI Assistants (No. arXiv: 2502.15237). arXiv. https://doi.org/10.48550/arXiv.2502.15237
22. Myers, A., Vargas, M., Aksoy, S. G., Joslyn, C., Wilson, B., & Grimes, T. (2025). Talking to GDELT Through Knowledge Graphs (No. arXiv: 2503.07584). arXiv. https://doi.org/10.48550/arXiv.2503.07584
23. Ni, B., Liu, Z., Wang, L., Lei, Y., Zhao, Y., Cheng, X., Zeng, Q., Dong, L., Xia, Y., Kenthapadi, K., Rossi, R., Dernoncourt, F., Tanjim, M. M., Ahmed, N., Liu, X., Fan, W., Blasch, E., Wang, Y., Jiang, M., & Derr, T. (2025). Towards Trustworthy Retrieval Augmented Generation for Large Language Models: A Survey (No. arXiv: 2502.06872). arXiv. https://doi.org/10.48550/arXiv.2502.06872
24. Singh, A., Ehtesham, A., Kumar, S., & Khoei, T. T. (2025). Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG (No. arXiv: 2501.09136). arXiv. https://doi.org/10.48550/arXiv.2501.09136
25. Wang, H., Feng, Y., Xie, X., & Zhou, S. K. (2025). Path Pooling: Train-Free Structure Enhancement for Efficient Knowledge Graph Retrieval-Augmented Generation (No. arXiv: 2503.05203). arXiv. https://doi.org/10.48550/arXiv.2503.05203
26. Wang, S., Fang, Y., Zhou, Y., Liu, X., & Ma, Y. (2025). ArchRAG: Attributed Community-based Hierarchical Retrieval-Augmented Generation (No. arXiv: 2502.09891). arXiv. https://doi.org/10.48550/arXiv.2502.09891
27. Yin, C., Wei, E., Zhang, Z., & Zhan, Z. (2025). PaperHelper: Knowledge-Based LLM QA Paper Reading Assistant (No. arXiv: 2502.14271). arXiv. https://doi.org/10.48550/arXiv.2502.14271
28. Yuan, X., Liu, Y., Di, S., Wu, S., Zheng, L., Meng, R., Chen, L., Zhou, X., & Yin, J. (2025). A Pilot Empirical Study on When and How to Use Knowledge Graphs as Retrieval Augmented Generation (No. arXiv: 2502.20854). arXiv. https://doi.org/10.48550/arXiv.2502.20854
29. Zhang, J., Liu, Y., Wang, W., Liu, Q., Wu, S., Wang, L., & Chua, T.-S. (2025). Personalized Text Generation with Contrastive Activation Steering (No. arXiv: 2503.05213). arXiv. https://doi.org/10.48550/arXiv.2503.05213
30. Zhang, Z., Feng, Y., & Zhang, M. (2025). LevelRAG: Enhancing Retrieval-Augmented Generation with Multi-hop Logic Planning over Rewriting Augmented Searchers (No. arXiv: 2502.18139). arXiv. https://doi.org/10.48550/arXiv.2502.18139
31. Zhao, J., Ji, Z., Fan, Z., Wang, H., Niu, S., Tang, B., Xiong, F., & Li, Z. (2025). MoC: Mixtures of Text Chunking Learners for Retrieval-Augmented Generation System (No. arXiv: 2503.09600). arXiv. https://doi.org/10.48550/arXiv.2503.09600
32. Zheng, Z., Ni, X., & Hong, P. (2025). Multiple Abstraction Level Retrieve Augment Generation (No. arXiv: 2501.16952). arXiv. https://doi.org/10.48550/arXiv.2501.16952
33. Zhou, J., & Chen, L. (2025). OpenRAG: Optimizing RAG End-to-End via In-Context Retrieval Learning (No. arXiv: 2503.08398). arXiv. https://doi.org/10.48550/arXiv.2503.08398
34. Zhou, Y., Su, Y., Sun, Y., Wang, S., Wang, T., He, R., Zhang, Y., Liang, S., Liu, X., Ma, Y., & Fang, Y. (2025). In-depth Analysis of Graph-based RAG in a Unified Framework (No. arXiv: 2503.04338). arXiv. https://doi.org/10.48550/arXiv.2503.04338
35. Zhu, X., Xie, Y., Liu, Y., Li, Y., & Hu, W. (2025). Knowledge Graph-Guided Retrieval Augmented Generation (No. arXiv: 2502.06864). arXiv. https://doi.org/10.48550/arXiv.2502.06864
36. Alonso, N., Figliolia, T., Ndirango, A., & Millidge, B. (2024). Toward Conversational Agents with Context and Time Sensitive Long-term Memory (No. arXiv: 2406.00057). arXiv. https://doi.org/10.48550/arXiv.2406.00057
37. Anokhin, P., Semenov, N., Sorokin, A., Evseev, D., Burtsev, M., & Burnaev, E. (2024). AriGraph: Learning Knowledge Graph World Models with Episodic Memory for LLM Agents (No. arXiv: 2407.04363). arXiv. https://doi.org/10.48550/arXiv.2407.04363
38. Chen, H., Pasunuru, R., Weston, J., & Celikyilmaz, A. (2023). Walking Down the Memory Maze: Beyond Context Limit through Interactive Reading (No. arXiv: 2310.05029). arXiv. https://doi.org/10.48550/arXiv.2310.05029
39. Chen, S., Zhao, Z., Zhao, Y., & Li, X. (2024). Apollonion: Profile-centric Dialog Agent (No. arXiv: 2404.08692). arXiv. https://doi.org/10.48550/arXiv.2404.08692
40. Gao, H., & Zhang, Y. (2024). Memory Sharing for Large Language Model based Agents (No. arXiv: 2404.09982). arXiv. https://doi.org/10.48550/arXiv.2404.09982
41. Guo, J., Li, N., Qi, J., Yang, H., Li, R., Feng, Y., Zhang, S., & Xu, M. (2024). Empowering Working Memory for Large Language Model Agents (No. arXiv: 2312.17259). arXiv. https://doi.org/10.48550/arXiv.2312.17259
42. Gutiérrez, B. J., Shu, Y., Gu, Y., Yasunaga, M., & Su, Y. (2025). HippoRAG: Neurobiologically Inspired Long-Term Memory for Large Language Models (No. arXiv: 2405.14831). arXiv. https://doi.org/10.48550/arXiv.2405.14831
43. Hou, Y., Tamoto, H., & Miyashita, H. (2024). 《My agent understands me better》: Integrating Dynamic Human-like Memory Recall and Consolidation in LLM-Based Agents. Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, 1–7. https://doi.org/10.1145/3613905.3650839
44. Hu, M., Chen, T., Chen, Q., Mu, Y., Shao, W., & Luo, P. (2024). HiAgent: Hierarchical Working Memory Management for Solving Long-Horizon Agent Tasks with Large Language Model (No. arXiv: 2408.09559). arXiv. https://doi.org/10.48550/arXiv.2408.09559
45. Hu, P., & Ying, X. (2025). Unified Mind Model: Reimagining Autonomous Agents in the LLM Era (No. arXiv: 2503.03459). arXiv. https://doi.org/10.48550/arXiv.2503.03459
46. Jiang, J., Zhou, K., Zhao, W. X., Song, Y., Zhu, C., Zhu, H., & Wen, J.-R. (2024). KG-Agent: An Efficient Autonomous Agent Framework for Complex Reasoning over Knowledge Graph (No. arXiv: 2402.11163). arXiv. https://doi.org/10.48550/arXiv.2402.11163
47. Jiang, X., Li, F., Zhao, H., Wang, J., Shao, J., Xu, S., Zhang, S., Chen, W., Tang, X., Chen, Y., Wu, M., Ma, W., Wang, M., & Chen, T. (2024). Long Term Memory: The Foundation of AI Self-Evolution (No. arXiv: 2410.15665). arXiv. https://doi.org/10.48550/arXiv.2410.15665
48. Kim, T., François-Lavet, V., & Cochez, M. (2024). Leveraging Knowledge Graph-Based Human-Like Memory Systems to Solve Partially Observable Markov Decision Processes (No. arXiv: 2408.05861). arXiv. https://doi.org/10.48550/arXiv.2408.05861
49. Li, H., Yang, C., Zhang, A., Deng, Y., Wang, X., & Chua, T.-S. (2025). Hello Again! LLM-powered Personalized Agent for Long-term Dialogue (No. arXiv: 2406.05925). arXiv. https://doi.org/10.48550/arXiv.2406.05925
50. Liang, X., Tao, M., Xia, Y., Shi, T., Wang, J., & Yang, J. (2024). Self-evolving Agents with reflective and memory-augmented abilities (No. arXiv: 2409.00872). arXiv. https://doi.org/10.48550/arXiv.2409.00872
51. Liu, J., Gu, S., Li, D., Zhang, G., Han, M., Gu, H., Zhang, P., Lu, T., Shang, L., & Gu, N. (2025). Enhancing Cross-Domain Recommendations with Memory-Optimized LLM-Based User Agents (No. arXiv: 2502.13843). arXiv. https://doi.org/10.48550/arXiv.2502.13843
52. Liu, L., Yang, X., Shen, Y., Hu, B., Zhang, Z., Gu, J., & Zhang, G. (2023). Think-in-Memory: Recalling and Post-thinking Enable LLMs with Long-Term Memory (No. arXiv: 2311.08719). arXiv. https://doi.org/10.48550/arXiv.2311.08719
53. Liu, N., Chen, L., Tian, X., Zou, W., Chen, K., & Cui, M. (2024). From LLM to Conversational Agent: A Memory Enhanced Architecture with Fine-Tuning of Large Language Models (No. arXiv: 2401.02777). arXiv. https://doi.org/10.48550/arXiv.2401.02777
54. Liu, W., Zhang, R., Zhou, A., Gao, F., & Liu, J. (2025). Echo: A Large Language Model with Temporal Episodic Memory (No. arXiv: 2502.16090). arXiv. https://doi.org/10.48550/arXiv.2502.16090
55. Maharana, A., Lee, D.-H., Tulyakov, S., Bansal, M., Barbieri, F., & Fang, Y. (2024). Evaluating Very Long-Term Conversational Memory of LLM Agents (No. arXiv: 2402.17753). arXiv. https://doi.org/10.48550/arXiv.2402.17753
56. McKee, K. L. (2025). Meta-Learning to Explore via Memory Density Feedback (No. arXiv: 2503.02831). arXiv. https://doi.org/10.48550/arXiv.2503.02831
57. Michelman, J., Baratalipour, N., & Abueg, M. (2025). Enhancing Reasoning with Collaboration and Memory (No. arXiv: 2503.05944). arXiv. https://doi.org/10.48550/arXiv.2503.05944
58. Mumuni, A., & Mumuni, F. (2025). Large language models for artificial general intelligence (AGI): A survey of foundational principles and approaches (No. arXiv: 2501.03151). arXiv. https://doi.org/10.48550/arXiv.2501.03151
59. Ong, K. T., Kim, N., Gwak, M., Chae, H., Kwon, T., Jo, Y., Hwang, S., Lee, D., & Yeo, J. (2025). Towards Lifelong Dialogue Agents via Timeline-based Memory Management (No. arXiv: 2406.10996). arXiv. https://doi.org/10.48550/arXiv.2406.10996
60. Pan, H., Zhai, Z., Yuan, H., Lv, Y., Fu, R., Liu, M., Wang, Z., & Qin, B. (2024). KwaiAgents: Generalized Information-seeking Agent System with Large Language Models (No. arXiv: 2312.04889). arXiv. https://doi.org/10.48550/arXiv.2312.04889
61. Pan, Z., Wu, Q., Jiang, H., Luo, X., Cheng, H., Li, D., Yang, Y., Lin, C.-Y., Zhao, H. V., Qiu, L., & Gao, J. (2025). On Memory Construction and Retrieval for Personalized Conversational Agents (No. arXiv: 2502.05589). arXiv. https://doi.org/10.48550/arXiv.2502.05589
62. Peng, Q., Liu, H., Huang, H., Yang, Q., & Shao, M. (2025). A Survey on LLM-powered Agents for Recommender Systems (No. arXiv: 2502.10050). arXiv. https://doi.org/10.48550/arXiv.2502.10050
63. Pink, M., Wu, Q., Vo, V. A., Turek, J., Mu, J., Huth, A., & Toneva, M. (2025). Position: Episodic Memory is the Missing Piece for Long-Term LLM Agents (No. arXiv: 2502.06975). arXiv. https://doi.org/10.48550/arXiv.2502.06975
64. Rappazzo, B. H., Wang, Y., Ferber, A., & Gomes, C. (2024). GEM-RAG: Graphical Eigen Memories For Retrieval Augmented Generation (No. arXiv: 2409.15566). arXiv. https://doi.org/10.48550/arXiv.2409.15566
65. Rasmussen, P., Paliychuk, P., Beauvais, T., Ryan, J., & Chalef, D. (2025). Zep: A Temporal Knowledge Graph Architecture for Agent Memory (No. arXiv: 2501.13956). arXiv. https://doi.org/10.48550/arXiv.2501.13956
66. Schmied, T., Paischer, F., Patil, V., Hofmarcher, M., Pascanu, R., & Hochreiter, S. (2024). Retrieval-Augmented Decision Transformer: External Memory for In-context RL (No. arXiv: 2410.07071). arXiv. https://doi.org/10.48550/arXiv.2410.07071
67. Shang, J., Zheng, Z., Wei, J., Ying, X., Tao, F., & Team, M. (2024). AI-native Memory: A Pathway from LLMs Towards AGI (No. arXiv: 2406.18312). arXiv. https://doi.org/10.48550/arXiv.2406.18312
68. Singh, A., Ehtesham, A., Kumar, S., & Khoei, T. T. (2025). Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG (No. arXiv: 2501.09136). arXiv. https://doi.org/10.48550/arXiv.2501.09136
69. Sumers, T. R., Yao, S., Narasimhan, K., & Griffiths, T. L. (2024). Cognitive Architectures for Language Agents (No. arXiv: 2309.02427). arXiv. https://doi.org/10.48550/arXiv.2309.02427
70. Sun, Y., Fu, H., Littman, M., & Konidaris, G. (2025). Knowledge Retention for Continual Model-Based Reinforcement Learning (No. arXiv: 2503.04256). arXiv. https://doi.org/10.48550/arXiv.2503.04256
71. Tan, J. C. M., Saroj, P., Runwal, B., Maheshwari, H., Sheng, B. L. Y., Cottrill, R., Chona, A., Kumar, A., & Motani, M. (2024). TaskGen: A Task-Based, Memory-Infused Agentic Framework using StrictJSON (No. arXiv: 2407.15734). arXiv. https://doi.org/10.48550/arXiv.2407.15734
72. Tan, Z., Yan, J., Hsu, I.-H., Han, R., Wang, Z., Le, L. T., Song, Y., Chen, Y., Palangi, H., Lee, G., Iyer, A., Chen, T., Liu, H., Lee, C.-Y., & Pfister, T. (2025). In Prospect and Retrospect: Reflective Memory Management for Long-term Personalized Dialogue Agents (No. arXiv: 2503.08026). arXiv. https://doi.org/10.48550/arXiv.2503.08026
73. Wang, P., Li, Z., Zhang, N., Xu, Z., Yao, Y., Jiang, Y., Xie, P., Huang, F., & Chen, H. (2024). WISE: Rethinking the Knowledge Memory for Lifelong Model Editing of Large Language Models (No. arXiv: 2405.14768). arXiv. https://doi.org/10.48550/arXiv.2405.14768
74. Wang, Q., Gao, Z., & Xu, R. (2023). Graph Agent: Explicit Reasoning Agent for Graphs (No. arXiv: 2310.16421). arXiv. https://doi.org/10.48550/arXiv.2310.16421
75. Wang, X., Wang, S., Zhu, Y., & Liu, B. (2025). R \(^3\) Mem: Bridging Memory Retention and Retrieval via Reversible Compression (No. arXiv: 2502.15957). arXiv. https://doi.org/10.48550/arXiv.2502.15957
76. Wang, Z. Z., Mao, J., Fried, D., & Neubig, G. (2024). Agent Workflow Memory (No. arXiv: 2409.07429). arXiv. https://doi.org/10.48550/arXiv.2409.07429
77. Wei, J., Ying, X., Gao, T., Bao, F., Tao, F., & Shang, J. (2025). AI-native Memory 2.0: Second Me (No. arXiv: 2503.08102). arXiv. https://doi.org/10.48550/arXiv.2503.08102
78. Xu, W., Liang, Z., Mei, K., Gao, H., Tan, J., & Zhang, Y. (2025). A-MEM: Agentic Memory for LLM Agents (No. arXiv: 2502.12110). arXiv. https://doi.org/10.48550/arXiv.2502.12110
79. Yan, X., Feng, S., Yuan, J., Xia, R., Wang, B., Zhang, B., & Bai, L. (2025). SurveyForge: On the Outline Heuristics, Memory-Driven Generation, and Multi-dimensional Evaluation for Automated Survey Writing (No. arXiv: 2503.04629). arXiv. https://doi.org/10.48550/arXiv.2503.04629
80. Yang, H., Lin, Z., Wang, W., Wu, H., Li, Z., Tang, B., Wei, W., Wang, J., Tang, Z., Song, S., Xi, C., Yu, Y., Chen, K., Xiong, F., Tang, L., & E, W. (2024). \(\text{Memory}^3\) : Language Modeling with Explicit Memory. Journal of Machine Learning, 3(3), 300–346. https://doi.org/10.4208/jml.240708
81. Yang, W., Li, Y., Fang, M., & Chen, L. (2025). MTPChat: A Multimodal Time-Aware Persona Dataset for Conversational Agents (No. arXiv: 2502.05887). arXiv. https://doi.org/10.48550/arXiv.2502.05887
82. Yue, S., Wang, S., Chen, W., Huang, X., & Wei, Z. (2025). Synergistic Multi-Agent Framework with Trajectory Learning for Knowledge-Intensive Tasks (No. arXiv: 2407.09893). arXiv. https://doi.org/10.48550/arXiv.2407.09893
83. Zeng, R., Fang, J., Liu, S., & Meng, Z. (2024). On the Structural Memory of LLM Agents (No. arXiv: 2412.15266). arXiv. https://doi.org/10.48550/arXiv.2412.15266
84. Zhang, Y., Chen, Z., Guo, L., Xu, Y., Zhang, W., & Chen, H. (2024). Making Large Language Models Perform Better in Knowledge Graph Completion (No. arXiv: 2310.06671). arXiv. https://doi.org/10.48550/arXiv.2310.06671
85. Zhang, Z., Bo, X., Ma, C., Li, R., Chen, X., Dai, Q., Zhu, J., Dong, Z., & Wen, J.-R. (2024). A Survey on the Memory Mechanism of Large Language Model based Agents (No. arXiv: 2404.13501). arXiv. https://doi.org/10.48550/arXiv.2404.13501
86. Zheng, J., Shi, C., Cai, X., Li, Q., Zhang, D., Li, C., Yu, D., & Ma, Q. (2025). Lifelong Learning of Large Language Model based Agents: A Roadmap (No. arXiv: 2501.07278). arXiv. https://doi.org/10.48550/arXiv.2501.07278
87. Zhong, W., Guo, L., Gao, Q., Ye, H., & Wang, Y. (2023). MemoryBank: Enhancing Large Language Models with Long-Term Memory (No. arXiv: 2305.10250). arXiv. https://doi.org/10.48550/arXiv.2305.10250
88. Bran, A. M., Cox, S., Schilter, O., Baldassari, C., White, A. D., & Schwaller, P. (2023). ChemCrow: Augmenting large-language models with chemistry tools (No. arXiv: 2304.05376). arXiv. https://doi.org/10.48550/arXiv.2304.05376
89. Chen, W., Su, Y., Zuo, J., Yang, C., Yuan, C., Chan, C.-M., Yu, H., Lu, Y., Hung, Y.-H., Qian, C., Qin, Y., Cong, X., Xie, R., Liu, Z., Sun, M., & Zhou, J. (2023). AgentVerse: Facilitating Multi-Agent Collaboration and Exploring Emergent Behaviors (No. arXiv: 2308.10848). arXiv. https://doi.org/10.48550/arXiv.2308.10848
90. Du, Y., Li, S., Torralba, A., Tenenbaum, J. B., & Mordatch, I. (2023). Improving Factuality and Reasoning in Language Models through Multiagent Debate (No. arXiv: 2305.14325). arXiv. https://doi.org/10.48550/arXiv.2305.14325
91. Ge, Y., Hua, W., Mei, K., Ji, J., Tan, J., Xu, S., Li, Z., & Zhang, Y. (2023). OpenAGI: When LLM Meets Domain Experts (No. arXiv: 2304.04370). arXiv. https://doi.org/10.48550/arXiv.2304.04370
92. Guo, J., Yang, B., Yoo, P., Lin, B. Y., Iwasawa, Y., & Matsuo, Y. (2024). Suspicion-Agent: Playing Imperfect Information Games with Theory of Mind Aware GPT-4 (No. arXiv: 2309.17277). arXiv. https://doi.org/10.48550/arXiv.2309.17277
93. Holt, S., Luyten, M. R., & Schaar, M. van der. (2024). L2MAC: Large Language Model Automatic Computer for Extensive Code Generation (No. arXiv: 2310.02003). arXiv. https://doi.org/10.48550/arXiv.2310.02003
94. Ifargan, T., Hafner, L., Kern, M., Alcalay, O., & Kishony, R. (2024). Autonomous LLM-driven research from data to human-verifiable research papers (No. arXiv: 2404.17605). arXiv. https://doi.org/10.48550/arXiv.2404.17605
95. Qian, C., Liu, W., Liu, H., Chen, N., Dang, Y., Li, J., Yang, C., Chen, W., Su, Y., Cong, X., Xu, J., Li, D., Liu, Z., & Sun, M. (2024). ChatDev: Communicative Agents for Software Development (No. arXiv: 2307.07924). arXiv. https://doi.org/10.48550/arXiv.2307.07924
96. Shen, Y., Song, K., Tan, X., Li, D., Lu, W., & Zhuang, Y. (2023). HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face (No. arXiv: 2303.17580). arXiv. https://doi.org/10.48550/arXiv.2303.17580
97. Singh, A., Ehtesham, A., Kumar, S., & Khoei, T. T. (2025). Agentic Retrieval-Augmented Generation: A Survey on Agentic RAG (No. arXiv: 2501.09136). arXiv. https://doi.org/10.48550/arXiv.2501.09136
98. Wang, G., Xie, Y., Jiang, Y., Mandlekar, A., Xiao, C., Zhu, Y., Fan, L., & Anandkumar, A. (2023). Voyager: An Open-Ended Embodied Agent with Large Language Models (No. arXiv: 2305.16291). arXiv. https://doi.org/10.48550/arXiv.2305.16291
99. Wang, L., Xu, W., Lan, Y., Hu, Z., Lan, Y., Lee, R. K.-W., & Lim, E.-P. (2023). Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models (No. arXiv: 2305.04091). arXiv. https://doi.org/10.48550/arXiv.2305.04091
100. Wu, Q., Bansal, G., Zhang, J., Wu, Y., Li, B., Zhu, E., Jiang, L., Zhang, X., Zhang, S., Liu, J., Awadallah, A. H., White, R. W., Burger, D., & Wang, C. (2023). AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation (No. arXiv: 2308.08155). arXiv. https://doi.org/10.48550/arXiv.2308.08155
101. Xie, T., Zhou, F., Cheng, Z., Shi, P., Weng, L., Liu, Y., Hua, T. J., Zhao, J., Liu, Q., Liu, C., Liu, L. Z., Xu, Y., Su, H., Shin, D., Xiong, C., & Yu, T. (2023). OpenAgents: An Open Platform for Language Agents in the Wild (No. arXiv: 2310.10634). arXiv. https://doi.org/10.48550/arXiv.2310.10634
102. Zhang, A., Chen, Y., Sheng, L., Wang, X., & Chua, T.-S. (2024). On Generative Agents in Recommendation (No. arXiv: 2310.10108). arXiv. https://doi.org/10.48550/arXiv.2310.10108
103. Zhou, W., Jiang, Y. E., Li, L., Wu, J., Wang, T., Qiu, S., Zhang, J., Chen, J., Wu, R., Wang, S., Zhu, S., Chen, J., Zhang, W., Tang, X., Zhang, N., Chen, H., Cui, P., & Sachan, M. (2023). Agents: An Open-source Framework for Autonomous Language Agents (No. arXiv: 2309.07870). arXiv. https://doi.org/10.48550/arXiv.2309.07870
104. Zhuge, M., Liu, H., Faccio, F., Ashley, D. R., Csordás, R., Gopalakrishnan, A., Hamdi, A., Hammoud, H. A. A. K., Herrmann, V., Irie, K., Kirsch, L., Li, B., Li, G., Liu, S., Mai, J., Piękos, P., Ramesh, A., Schlag, I., Shi, W., … Schmidhuber, J. (2023). Mindstorms in Natural Language-Based Societies of Mind (No. arXiv: 2305.17066). arXiv. https://doi.org/10.48550/arXiv.2305.17066

RAG¶

教程汇总 ¶

• GitHub - NirDiamant/RAG_Techniques: This repository showcases various advanced techniques for Retrieval-Augmented Generation (RAG) systems. RAG systems combine information retrieval with generative models to provide accurate and contextually rich responses.
• GitHub - bRAGAI/bRAG-langchain: Everything you need to know to build your own RAG application
• GitHub - D-Star-AI/dsRAG: High-performance retrieval engine for unstructured data

KRAG¶

• GitHub - microsoft/SmartKG: This project accepts excel files as input which contains the description of a Knowledge Graph (Vertexes and Edges) and convert it into an in-memory Graph Store. This project implements APIs to search/filter/get nodes and relations from the in-memory Knowledge Graph. This project also provides a dialog management framework and enable a chatbot based on its knowledge graph.
• GitHub - CheMiguel23/MemoryMesh: A knowledge graph server that uses the Model Context Protocol (MCP) to provide structured memory persistence for AI models. v0.2.8
• GitHub - shaneholloman/mcp-knowledge-graph: MCP server enabling persistent memory for Claude through a local knowledge graph - fork focused on local development

Agent Document¶

• Blog | Letta
• The Letta Platform — Letta
• LangGraph
• 🦜️🔗 LangChain
• Get started with LangSmith | 🦜️🛠️ LangSmith

AI 代理关键技术方案总结 ¶

内存管理方案 ¶

1. 长短期记忆结构 - 长期记忆：通常通过向量数据库（如 Pinecone、Chroma、Milvus）实现，使用语义搜索检索相关信息
   • 应用实例：Pinecone 在 Cisco 的企业 AI 助手中用于准确、安全地搜索数百万文档
   • 应用实例：BabyAGI 使用 Pinecone 存储任务执行历史，即使关闭后也能保持记忆
   • 短期记忆（工作记忆）：由 LLM 维护和更新，用于当前对话或任务上下文
   • 应用实例：LangChain 的 ConversationBufferMemory 保存完整对话历史
   • 应用实例：AutoGen 中的代理可以在对话期间维护短期上下文
2. 记忆持久化方案 - 完整状态序列化：将代理的完整状态（包括记忆和工具状态）保存到文件或 Python 对象中
   • 应用实例：Letta（前身为 MemGPT）可以将代理状态序列化，无需外部数据库
   • 应用实例：LangGraph 的 Memory Store 提供低级抽象，让用户完全控制代理记忆
   • 会话持久记忆：确保数据在多个会话之间保存
   • 应用实例：Letta 的核心记忆功能允许代理记住用户信息（如名字），即使在会话结束后
   • 应用实例：LangChain 的 ConversationEntityMemory 可以跟踪对话中提到的实体
   • 无限记忆回忆：如 BabyAGI 的一些变体，可以回忆 " 无限 " 记忆，即使在关闭后也不会丢失记忆
   • 应用实例：BabyAGI 的变体使用 Pinecone 向量数据库和记忆计数器保存索引位置
   • 应用实例：MemGPT 使用多层记忆架构，允许无限制的记忆容量
3. 记忆管理技术 - 自编辑记忆：允许聊天机器人自我编辑其记忆
   • 应用实例：Letta 的 core_memory_replace 功能允许代理更新核心记忆（如用户名字）
   • 应用实例：MemGPT 的 store_message 功能允许代理存储重要的用户细节
   • 分层记忆管理：智能管理不同层级的记忆，在 LLM 有限的上下文窗口内有效提供扩展上下文
   • 应用实例：MemGPT 的三层记忆架构：核心记忆、回忆记忆和归档记忆
   • 应用实例：LangChain 的 ConversationSummaryBufferMemory 通过总结压缩长对话
   • 记忆总结：通过总结压缩长期记忆，保留关键信息
   • 应用实例：LangChain 的 ConversationSummaryMemory 创建对话摘要而非存储完整历史
   • 应用实例：MemGPT 可以将对话总结为可重用的记忆
4. 记忆检索策略 - 相关性查询：AI 在其记忆中查找与当前查询相关的记忆和过去的查询
   • 应用实例：MemGPT 的 conversation_search 功能允许搜索整个消息历史
   • 应用实例：LangChain 的 VectorStoreRetrieverMemory 使用向量相似性检索相关记忆
   • 记忆索引：使用计数器保存索引位置，便于高效检索
   • 应用实例：BabyAGI 变体使用记忆计数器保存索引位置
   • 应用实例：Pinecone 的索引功能支持高效的向量搜索
5. 记忆类型与实现框架 - 对话记忆：存储和检索对话历史
   • 应用实例：LangChain 提供多种对话记忆类型：Buffer、Summary、Entity、KnowledgeGraph
   • 应用实例：LangGraph 的 ReAct Memory Agent 可以保存用户偏好，跨对话线程使用
   • 向量记忆：使用嵌入向量存储和检索信息
   • 应用实例：Pinecone 用于 RAG（检索增强生成）应用，提供准确的知识检索
   • 应用实例：Milvus Lite 用于为 LangChain 代理提供长期记忆
   • 组织记忆：为团队协作设计的记忆系统
   • 应用实例：某些闭源项目专注于组织记忆和团队协作
   • 应用实例：LangChain 的 Memory Store 可以根据用户 ID 范围化记忆
6. 知识图谱记忆 - 图结构表示：使用节点和边表示实体及其关系，提供结构化的知识表示
   • 应用实例：FalkorDB 提供超低延迟图数据库解决方案，优化 AI 代理的知识存储
   • 应用实例：LangChain 的 ConversationKnowledgeGraphMemory 构建对话中实体的关系图
   • 多步推理能力：通过遍历关系图进行复杂推理和决策
   • 应用实例：Zep 的时间知识图谱可以跟踪事实如何随时间变化
   • 应用实例：图数据库支持复杂查询，使代理能够执行复杂分析并产生更好的结果
   • 语义关系保存：保持实体间的复杂语义关系，而不仅仅是简单的键值对
   • 应用实例：SAP 知识图谱通过连接 SAP 数据与业务上下文，释放数据全部价值
   • 应用实例：Zep 的知识图谱智能融合聊天消息和业务数据
7. 实体记忆 - 实体提取与总结：从对话中提取命名实体并生成摘要
   • 应用实例：LangChain 的 ConversationEntityMemory 从最近的聊天历史中提取命名实体
   • 应用实例：实体记忆可以跟踪对话中提到的人物、地点、组织等
   • 实体存储：使用可交换的实体存储，在对话间持久化实体
   • 应用实例：LangChain 支持多种实体存储：内存、Redis、SQLite 等
   • 应用实例：实体记忆可以记住 "Sam 是 Daimon 公司的创始人 " 等关键事实
   • 实体更新：随着对话进行，不断更新和丰富实体信息
   • 应用实例：当获取新信息时，实体记忆会更新现有实体的描述
   • 应用实例：Zep 标记过时的事实为无效，保持实体信息的最新状态
8. 用户画像记忆 - 用户偏好跟踪：记录用户的偏好、兴趣和行为模式
   • 应用实例：LangGraph 的 Memory Store 可以根据用户 ID 存储用户特定的记忆
   • 应用实例：Zep 可以为每个用户构建个性化的知识图谱
   • 个性化响应生成：基于用户画像定制回复
   • 应用实例：高度个性化的 AI 助手可以记住所有用户偏好和之前的交互
   • 应用实例：Zep 的时间推理能力使代理能够理解用户状态的变化
   • 跨会话用户识别：在多个会话中识别和记住同一用户
   • 应用实例：Letta 可以在用户再次登录时记住对话细节
   • 应用实例：企业级记忆系统支持 SOC 2 Type II 合规和隐私控制

多代理协作方案 ¶

1. 控制器架构 - 动态决策控制器：使用 LLM 动态决定下一个执行动作的代理，考虑先前的动作、环境和当前状态的目标 - 符号控制：使用标准操作流程 (SOP) 定义整体任务的子目标 / 子任务
2. 代理角色分配 - 专家代理：每个代理扮演特定专业领域的专家角色 - 角色扮演代理：代理具有特定角色、背景故事、目标和记忆 - 人机交互：框架支持人类用户扮演代理角色，输入自己的动作，与环境中的其他语言代理交互
3. 协作模式 - 对话式协作：代理通过对话交流信息和想法 - 任务分解协作：将复杂任务分解为子任务，由不同代理处理 - 层次协作：代表代理间协作的层次结构
4. 多 LLM 系统 - 混合强度模型：对需要强推理和指令遵循的代理使用更强大的 LLM，将简单任务委托给较弱 / 本地 LLM - 自动图优化器：优化节点级 LLM 提示和改进代理编排

工具使用方案 ¶

1. 工具集成技术 - 函数调用：通过 OpenAI 的 function-calling 或类似机制集成外部工具 - 自定义工具 /API：开发者可以添加自定义工具和 API - 工具消息：支持与函数调用等效的原生 ToolMessage
2. 工具类型 - 搜索工具：结合搜索、抓取、分块和提取功能 - 迷你代理工具：将小型专用代理作为工具使用 - 命令行工具：执行 shell 命令的能力 - 专业领域工具：如 ChemCrow 集成的 13 种专家设计工具，增强 LLM 在化学领域的性能
3. 工具使用模式 - ReAct 模式：思考、行动、行动输入、观察的循环模式 - 工具代理：为子任务提供最佳行动系列的专用代理 - 工具设计与调试：能够设计、编码和调试自己的工具
4. 规划与执行 - Plan-and-Solve 方法：通过大型语言模型改进零样本链式思考推理 - 任务规划：使用 ChatGPT 分析用户请求，理解意图并分解为可解决的任务 - 工作流步骤批准：在相关工作流步骤请求批准，确保执行预期操作