Subliminal Signals in Preference Labels
As AI systems approach superhuman capabilities, scalable oversight increasingly relies on LLM-as-a-judge frameworks where models evaluate and guide each other's training. A core assumption is that binary preference labels provide only semantic supervision about response quality. We challenge this assumption by demonstrating that preference labels can function as a covert communication channel. We show that even when a neutral student model generates semantically unbiased completions, a biased judge can transmit unintended behavioral traits through preference assignments, which even strengthen across iterative alignment rounds. Our findings suggest that robust oversight in superalignment s...
| Authors | Isotta Magistrali, Frédéric Berdoz, Sam Dauncey et al. |
|---|---|
| Primary category | cs.LG |
| Published | 1 March 2026 |
| Last updated at source | 12 March 2026 |
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.