Skip to main content
Search

Hashmath Shaik

1 related articles

Subscribe to updates
  1. arXiv

    Pre-verdict residual stream activations predict LLM judge position flips with AUROC .621–.850 and transfer to .685–.853

    Using nested grouped cross-validation on 534 JudgeBench pairs and 1,802 MT-Bench comparisons, this work trains L2-regularized linear probes on residual stream activations recorded immediately before the verdict to predict position flips, where swapping candidate order changes the judge's selected response identity; AUROCs reach .621–.850 across four judges, exceeding a combined baseline of verbalized confidence, verdict logits, length differences, and the initial choice by .062–.113, and frozen probes transfer to MT-Bench at .685–.853.

You’re all caught up