כתבה
arXiv cs.AI ·
SiLR: Structure-Preserving Admission and Process Reward for LLM Tool Agents
תקציר מקורי באנגליתarXiv:2609.04629v1 Announce Type: new Abstract: A runtime gate for an LLM tool agent is usually cast as a filter. In a ReAct loop a rejected proposal is followed by another at the same state, so the gate is a search operator over the proposal stream whose admission criterion shapes which trajectories are reachable. We study post-violation recovery admission, where progress must be admitted while the system is still in violation, and identify the scalar projection trap: an aggregate-score gate accepts a locally improving proposal and commits the trajectory to a plateau. SiLR instead shadow-executes each proposal and admits it under a product order over the branch-level violation state (overloaded-branch support and per-branch severity). We prove that no scalar surrogate is sound for this or
קרא במקור המקורי
arxiv.org
פתח כתבה מקורית