Junhao Dong

We propose Equivalent Transformation Mechanism with KKT-Multiplier Regularization for solving SemiUOT and UOT

We propose the first robust weak-to-strong generalization framework to elicit robust knowledge from a strong student VLM in an unsupervised scheme.

VLM unlearning by a closed-form ensemble of infinite number of functions with parameters uniformly sampled from a task arithmetic simplex.

How to employ the coherence between layers to design a PEFT method.