Audit at the Precision You Deploy: Quantization Reshapes a Subliminally Learned Preference
arXiv:2611.00665v1 Announce Type: new Abstract: Subliminal learning allows AI models to inherit hidden preferences from seemingly unrelated training data, such as number sequences. Safety audits may evaluate models at standard 16-bit precision, while deployments may use compressed, quantized weights
Der Artikeltext wird beim nächsten Sync nachgeladen. Original bei arXiv cs.CL lesen
Quelle: arXiv cs.CL (arXiv cs.CL). Original öffnen