Category: LLM Agents
- Why Agent Training Always Crashes on Long-Horizon Tasks
- An Excellent New Systematic Survey of Self-Evolving Agents
- SJTU's 54-Page Review Illuminates the Evolutionary Path of Agent Cognitive Externalization
- Sharing Two Latest Harness Papers: One from Google, One from Microsoft
- A Unified Review of Agents: Harness, Memory, Skills, and Protocols
- How to Solve Context Explosion? Let Agents Actively 'Forget' Like Biological Organisms