arXiv preprint ·
Defence against the dark arts of distillation: Unsupervised discovery of subliminal learning
- Published
- Source
- arXiv preprint, posted before peer review
- Original
- Read it at arXiv
- Topics
Abstract
When a language model is distilled from a teacher that shares its base parameters, the student can inherit behavioural traits of the teacher---including misalignment---even from data that is semantically unrelated to the trait (Cloud et al., 2026). All existing defences require knowing the trait. We remove this requirement and prove that a subliminally transferred trait can be recovered from the parameter updates alone, without supervision, by exploiting a single asymmetry: the trait is expressed consistently across diverse datasets generated by the teacher, whereas task-specific and shared capabilities vary. The resulting estimator returns a weight-space direction that can be amplified to discover and interpret the trait or projected out to suppress it. Applied to the subliminal-learning benchmarks of (Cloud et al., 2026), our method recovers the transmitted animal-preference traits without any trait labels and suppresses their expression with almost no impact to task performance.
The authors’ abstract, as arXiv lists it. arXiv’s listings, abstracts included, are free for anyone to reuse.
Read it on ReadSearch
Each opens after you create an account, and brings you back to this paper.
More in Machine Learning (Statistics)
Scalable Learning of Determinantal Point Processes with Kernel-Structured Interactions
Prior-Informed Generalizable Low-Rank Representation Learning via Regularized Adversarial Reweighting
Two-way Guidance: Simultaneous Learning of Summary Statistics for Generative Bayes
ByteMark: A Unified Byte-Level Watermark for Detection Across Language Models
Universal Bounds for Self-Normalized Kullback-Leibler Estimation
Get the few papers worth your time in Machine Learning (Statistics) every morning.
ReadSearch reads the new papers in your field each day and brings you the few worth reading, each with its key finding in a sentence.