Agent Plasticity: Measuring Self-Improvement Through Experience
Source
arxiv.org
Author
Harman Singh, Anton Bakhtin, Rulin Shao, Gabriel Synnaeve, Ilia Kulikov, Rob Fergus, Sanjeev Arora, Kurt Keutzer, Jason Weston, Anuj Mahajan, Anirudh Goyal
Date
Why it matters
Fixed-point benchmarks miss whether an AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → actually learns from experience. This gives a way to measure learning efficiency and locate where self-improvement loops break down.
Terms in this piece · Glossary
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
articleSelf-Evolving Embodied Agents via Skill-Harness EvolutionPeidong Wang, Zhiming Ma, Ying Chang, Xufang Luo, Xiaocui Yang, Shi Feng, Yuqing Yang, Dongsheng Li
articleDo Agents Know When They Succeed? Calibrating Agent Confidence from Internal RepresentationsPriyanka Mary Mammen, Emil Joswin, Srujananjali Medicherla
articleSubagents vs Agent Skills: Executing Reusable Knowledge for Long-Horizon Agentic TasksWasu Top Piriyakulkij, Rachel Lawrence, Alicia Curth, Sushrut Karmalkar, Niranjani Prasad