1
Diagnosis Is Not Prescription: Linguistic Co-Adaptation Explains Patching Hazards in LLM Pipelines
从语言共适应角度揭示大模型流水线补丁的潜在风险，为AI安全评估提供关键新视角
arXiv:2605.21958v1 Announce Type: new Abstract: When a multi-module LLM agent fails, the module most responsible for the failure is not necessarily th…