1 paper · 1 filter
Yoon Jeonghun, Kim Dongchan
When a multi-module LLM agent fails, the module most responsible for the failure is not necessarily the best place to intervene. We demonstrate this Diagnostic Paradox empirically:…