Agent KB: A Hierarchical Memory Framework for Cross-Domain Agentic Problem Solving
Published in ICML 2025 Workshop on Collaborative and Federated Agentic Workflows, 2025
This work presents Agent KB, a hierarchical memory framework that enables intelligent agents to leverage knowledge from diverse domains to solve complex problems more effectively.
Key Contributions:
- Developed a hierarchical memory framework for agentic systems
- Enabled cross-domain knowledge transfer and experience reuse
- Demonstrated improved problem-solving capabilities across multiple domains
Research Area: Large Language Models, AI Agents, Knowledge Management
Recommended citation: X. Tang, T. Qin, T. Peng, Z. Zhou, D. Shao, T. Du, X. Wei, H. Zhu, G. Zhang, et al. (2025). "Agent KB: A Hierarchical Memory Framework for Cross-Domain Agentic Problem Solving." ICML 2025 Workshop on Collaborative and Federated Agentic Workflows.