LATIDIA · Investigación

Phronix: el paradigma de aprendizaje centrado en las capacidades para agentes

arXiv: 2610.31789v1Tipo de anuncio: nuevo Resumen: La informática científica se basa cada vez más en agentes basados en modelos de lenguaje grandes para tareas de ingeniería de rendimiento como instrumentación, análisis y portabilidad de flujos de trabajo. Cómo

La noticia

arXiv:2610.31789v1 Announce Type: new Abstract: Scientific computing increasingly relies on large language model-based agents for performance engineering tasks such as instrumentation, analysis, and workflow porting. However, agent effectiveness depends heavily on the surrounding harness, which typically relies on manually written skills and curated memory. This limits autonomous learning, forcing agents to repeatedly rediscover solutions rather than retain and reuse knowledge from prior executions. We introduce Phronix, a capability-centric learning paradigm that enables agentic harnesses to acquire, validate, and refine reusable capabilities from their own executions. Phronix operates through a predictor-validator-corrector loop: MCP tools and associated skills propose solutions, tool executions validate them, and observed outcomes refine the stored capabilities. It