כתבה
arXiv cs.AI ·
Sentry: שיקום מכישלונות של סוכנים LLM
Sentry: Learning to Recover from LLM Agent Failures at Test Time
Sentry הוא שכבת ניהול כישלונות שמאפשרת לסוכנים LLM להתאושש מכישלונות. היא עובדת עם מאגר לקחים חיצוני כדי להדריך שיקום. Sentry משפרת ביצועים ב-37% בממוצע.
תקציר מקורי באנגליתarXiv:2610.02994v1 Announce Type: cross Abstract: LLM agents often fail mid-task due to invalid tool calls, repeated actions, or poorly grounded reasoning, and learning from these failures is a path to reliability. We find that how failure knowledge reaches the agent matters as much as what it contains. Failure lessons are conditional: kept in the agent's context, they misfire when their failure is absent, and removing them from an evolving playbook improves performance. Runtime interventions, in contrast, act only when a failure occurs but do not learn from their repairs. We argue that failure knowledge is conditional knowledge and should be conditionally exposed, and instantiate this principle in Sentry, a failure-management layer that runs alongside the agent. When Sentry detects a fail
קרא במקור המקורי
arxiv.org
פתח כתבה מקורית