Jagged Judges: Epistemic Stability Under Silence, Pressure, and Persistence
arXiv cs.AI1mo4 min read
arXiv:2608.12645v1 Announce Type: new Abstract: LLM judges have become central infrastructure for model evaluations, online grading, and reward modeling. Judges are typically validated by accuracy on golden data, but accuracy says little about whether they are stable under re-prompting, challenge, or sustained pushback. We introduce the \emph{Wiggle Framework}, a unified stress test for epistemic stability in LLM judges. The framework decomposes judge robustness along three dimensions: Mechanical Consistency (stability under re-prompting and reframing), Single-turn Conviction (stability under