LATIDIA · Ciberseguridad

StateGate: Un método de optimización de avisos con fidelidad de comportamiento para agentes LLM de cadena larga

arXiv:2611.01585v1 Anuncio Tipo: nuevo Resumen: Actualmente, los agentes LLM de cadena larga suelen sufrir degradación del comportamiento en tareas complejas, donde las indicaciones se escriben completas pero los pasos no se ejecutan por completo. Para añadir

La noticia

arXiv:2611.01585v1 Announce Type: new Abstract: Currently, long-chain LLM agents commonly suffer from behavioral degradation in complex tasks, where prompts are written in full but steps are not executed in full. To address this issue, existing approaches face three major challenges. First, constraints are strongly bound to their textual carriers; once natural-language constraint sentences are removed, their binding force vanishes. Second, the execution process lacks independent verification, as the evidence for judgment is often supplied by the executing party itself, leading to the failure of self-referential validation. Third, existing remediation schemes are difficult to generalize, and hard-coded mechanisms cannot be transferred across tasks. In this paper, we propose StateGate, a novel