Skip to contentSkip to filters

We use cookies for analytics.

Exact Unlearning of Finetuning Data via Model Merging at Scale

Kevin Kuo

Carnegie Mellon University

Recording

26 min

Abstract

Kevin Kuo presents SIFT-Masks, a model-merging approach for exactly removing contributions from fine-tuned models. Local masks preserve task-specific utility, while a common sign constraint lets the masks be computed independently. Experiments merging up to 500 models compare accuracy with naive merging and measure the computational cost of exact removal against other merging methods.

Topics