PAC-Private Autoregressive Generation: Calibrating Noise to Ensemble Disagreement
Quick summary
arXiv:2609.05676v1 Announce Type: cross Abstract: Language models adapted on private text are often served through APIs, so privacy leakage occurs through generated outputs rather than exposed weights. Private prediction protects these releases. Methods such as PMixED incur privacy cost at each release and increasingly rely on the public model over long horizons. PAC privacy instead calibrates noise to output variability across possible secrets, adding less noise when predictions are stable. To our knowledge, PAC-private prediction has not previously been extended from classification to autore
Key takeaways
- arXiv:2609.05676v1 Announce Type: cross Abstract: Language models adapted on private text are often served through APIs, so privacy leakage occurs through generated outputs rather than exposed weights.
- Private prediction protects these releases.
- Methods such as PMixED incur privacy cost at each release and increasingly rely on the public model over long horizons.
Why it matters
The significance is not only the legal text but how it changes product design. Decisions around “PAC-Private Autoregressive Generation: Calibrating Noise to Ensemble Disagreement” may reshape data collection, model training, output accountability and market access.
Member comments