When Does a Language Model Commit? A Finite-Answer Theory of Pre-Verbalization Commitment

مقتطف من المصدر

تنظم هذه الصفحة محتوى عن الذكاء الاصطناعي والبحث لتسهيل قراءته والاستشهاد به والتحقق منه من قبل الباحثين والطلاب وأدوات الذكاء الاصطناعي.

arXiv:2605.06723v1 Announce Type: new Abstract: Language models often generate reasoning before giving a final answer, but the visible answer does not reveal when the model's answer preference became stable. We study this question through a narrow computable object: \emph{finite-answer preference stabilization}. For a model state and specified answer verbalizers, we project the model's own continuation probabilities onto a finite answer set; in binary tasks this yields an exact log-odds code, $\delta(\xi)=S_\theta(\mathrm{yes}\mid\xi)-S_\theta(\mathrm{no}\mid\xi)$. This target defines parser-based answer onset, retrospective stabilization time, and lead without relying on greedy rollouts or learned probes. In controlled delayed-verdict tasks with Qwen3-4B-Instruct, the contextual finite-answer projection stabilizes before the answer is parseable, with 17--31 token mean lead in the main templates and positive, shorter lead in a parser-clean replication. The signal tracks the model's eventual output rather than truth, is linearly recoverable from compact hidden summaries, is partly separable from cursor progress, and transfers as shared information without a single invariant coordinate. Diagnostics separate the measurement from online stopping, verbalizer-free belief, and causal answer control; exact steering shows local sensitivity of $\delta$ but not reliable generation control.

عرض المصدر الأصلي

المصدر الرسمي أو الأصلي: arXiv AI. تحقق دائما من التفاصيل في المصدر الأساسي.

العودة إلى الذكاء الاصطناعي والبحث