LATIDIA · Investigación

Agentes confiables de Long-Horizon LLM: un estudio empírico centrado en la trayectoria de dónde entran los errores y qué controles los alcanzan

arXiv:2611.13254v1 Announce Type: new Resumen: Cuando un agente LLM falla a mitad de una tarea larga, el análisis habitual registra un síntoma: una etiqueta de modo de falla, o el agente y el paso considerados responsables. No dice

La noticia

arXiv:2611.13254v1 Announce Type: new Abstract: When an LLM agent fails part-way through a long task, the usual analysis records a symptom: a failure-mode label, or the agent and step judged responsible. It does not say where the error entered, which state carried it to the symptom, whether anything in the trajectory contradicted it, or which control could have caught it. We define an episode record that ties a candidate fault to its origin, propagation, detection, response and consequence, located by operational phase and state domain, and that names the control families whose observation and reach cover it. We turned the record into a written annotation protocol and applied it to