의료 에이전트의 자가 진화 스킬 메모리로 일반화 추론 능력 향상
Experience Makes Skillful: Enabling Generalizable Medical Agent Reasoning via Self-Evolving Skill Memory
- 매체
- Hugging Face
- 게시
- 중요도
- 3/5
- 화제 지수
- 5
한국어 요약
의료 에이전트 시스템의 장기 임상 추론을 위해 자가 진화하는 스킬 기반 메모리 프레임워크 SkeMex를 제안했다. 상호작용 궤적을 구조화된 스킬로 증류하고 다층 저장소로 조직화하며, 환경 피드백으로부터 맥락 의존적 유용성을 추정해 메모리 검색과 관리를 수행한다. 다양한 임상 과제에서 기존 메모리 기반 에이전트를 능가하는 성능을 보였다.
원문 발췌 · English
Medical agent systems are increasingly expected to support interactive clinical decision making rather than only static question answering. In such settings, effective agents must reuse prior experience across evolving cases, yet existing memory mechanisms often retain raw historical traces that are redundant, noisy, and difficult to govern. More importantly, they rarely distinguish which memories are truly useful for future reasoning. This limits their ability to accumulate compact and reliable experience for long-horizon clinical reasoning. To close this gap, we propose SkeMex, a post-deployment self-evolution framework that improves medical agents through a skill-based memory without updating model weights. SkeMex distills informative interaction trajectories into structured skills that encode reusable procedural knowledge, and organizes them into a multi-branch repository spanning general, task-specific, and action-level experience. To determine which memories should be reused and retained, SkeMex estimates context-dependent utility from environment feedback and uses it to guide value-aware retrieval and repository governance. A closed-loop ``Read--Write--Assess--Govern" lifecycle further supports continual evolution by writing new skills, updating utilities, promoting useful memories, and removing harmful entries. Experiments across diverse clinical tasks show that SkeMex consistently outperforms representative memory-based agents in both offline and online settings. It also generalizes across model backbones and supports transferable skill memory. All data and code will be released publicly.