LATIDIA · Investigación

Humanizar a los agentes de MLE: Alternancia de sesión multiagente más allá de los límites de un solo agente

arXiv: 2610.13494v1Tipo de anuncio: nuevo Resumen: Los enfoques de agentes basados en LLM existentes para la ingeniería de aprendizaje automático (MLE) de horizonte largo a menudo utilizan ingeniería de contexto, implementada a través de arneses personalizados que administran herramientas

La noticia

arXiv:2610.13494v1 Announce Type: new Abstract: Existing LLM-based agent approaches for long-horizon machine learning engineering (MLE) often use context engineering, implemented through custom harnesses that manage tools, skills, and accumulated experience. However, we find that as single-session execution lengthens, the agent's reasoning intensity declines alongside slowing progress, summarized as long-horizon deceleration, and the benefits of these custom harnesses vanish as the underlying model improves, summarized as harness staleness. These findings suggest that their benefits may not persist across versions, which motivates a workflow-level approach in which agents build on one another's work across sessions. We model these alternating agent sessions as a context-reset semi-Markov decision process (CR-SMDP) and establish sufficient conditions