Beyond Bias Scores: Answerability Confounds in Activation Steering with Gradient-Constrained Direction Design (GCD).
arXiv:2610.17641v1 Announce Type: new Abstract: Bias scores in language-model evaluations often conflate what a model prefers with whether it answers at all. We identify this confound in \textsc{BBQ}, whose ambiguous score factorizes into answerability and sig
arXiv cs.AIResearch source·