GRAM selectively trains auxiliary modules so that ablating one at inference removes a targeted capability while preserving the rest, closely tracking data-filtered models at 5x lower cost across 5 capability profiles.
Advances in Neural Information Processing Systems , volume=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
ScaleErasure erases unsafe concepts in next-scale AR image generation via minimal logit guidance from unsafe and safe conditioned forward passes.
citing papers explorer
-
Modular Pretraining Enables Access Control
GRAM selectively trains auxiliary modules so that ablating one at inference removes a targeted capability while preserving the rest, closely tracking data-filtered models at 5x lower cost across 5 capability profiles.
-
ScaleErasure: Inference-Time Minimal Intervention for Precise Concept Erasure in Next-Scale Autoregressive Image Generation
ScaleErasure erases unsafe concepts in next-scale AR image generation via minimal logit guidance from unsafe and safe conditioned forward passes.