When and Why Do Linear Bias Probes Fail? A Geometric and Statistical Theory of Bias Detectability in Large Language Model Representations

arXiv:2609.22337v1 Announce Type: new Abstract: Linear probing is the standard instrument for detecting social biases in the hidden representations of large language models. Yet reported probe accuracies come almost exclusively from \emph{counterfactual} evaluations in which every input carries an…

Read the full story at arXiv stat.ML ↗

Timeline · 1 report

1. 2026-09-22 04:00 · arXiv stat.ML
   When and Why Do Linear Bias Probes Fail? A Geometric and Statistical Theory of Bias Detectability in Large Language Model Representations