Skip to content
arXiv cs.CL · Papers

Diagnosing Correctness Probes under Self-Judgement Confounding

arXiv:2607.16799v1 Announce Type: new Abstract: Hidden-state readouts can predict whether language-model outputs are correct, but objective correctness (OC) usually agrees with the model's own self-judgement (SJ), leaving the decoded signal semantically ambiguous. We construct conflict cases in which OC and SJ predict