Problem
Research questions and friction points this paper is trying to address.
Hallucination
Hidden-state Probes
Mean-shift
The Hallucination Signal Is a Mean Shift: Why Simple Probes Suffice
๐ค AI Summary
็ ็ฉถ้่ฟ็ฎๅๆข้ๆนๆณๆๆๆฃๆตๅคง่งๆจก่ฏญ่จๆจกๅ็ๅนป่ง้ฎ้ข๏ผๅ็ฐไฟกๅทไธป่ฆ็ฑๅไธๅๅผๅ็งป็ปๆ๏ผ็ฎๅไบๅคๆๆถๆ็้ๆฑใ
๐ Abstract
Hidden-state probes effectively detect LLM hallucinations, but the geometry of the signal remains poorly characterized, driving increasingly complex probe architectures. Across three 7B-scale models and three datasets in a paired-example paradigm, we find the signal overwhelmingly dominated by a single mean-shift component, and removing this direction collapses detection to chance. Shrinkage linear discriminant analysis closes about 73% of the gap between 1D and full-dimensional classifiers, so apparent architectural complexity largely reflects high-dimensional covariance estimation difficulty rather than exploitable non-linearity. A simple L2-regularized logistic regression (0.952 AUROC) bounds or outperforms twelve controlled architectural alternatives, and our multi-layer aggregation exceeds CLAP cross-layer attention probing under matched paradigm. Because the signal spans a contiguous layer band, LayerMix aggregates it to match oracle-layer performance without oracle access. Our claims characterize the geometry within the controlled paired-example paradigm. Our code is available at https://github.com/js-lee-AI/LayerMix.
Innovation
Methods, ideas, or system contributions that make the work stand out.
mean-shift
L2-regularized logistic regression
LayerMix
๐ Similar Papers
No similar papers found.
