Sanghyun Hong

IF-Guide: Influence Function-Guided Detoxification of LLMs

Evaluating Privacy Risks of Parameter-Efficient Fine-Tuning

Certified Robustness to Clean-label Poisoning Using Diffusion Denoising

Harnessing Input-adaptive Inference for Efficient Vision-and-Language Navigation

Hessian-Aware Training for Enhancing Model Resilience for In-Memory Computing

PAC-FNO: Parallel-Structured All-Component Fourier Neural Operators for Recognizing Low-Quality Images

Privacy Backdoors: Enhancing Membership Inference through Poisoning Pre-trained Models

Promoting Sparsity in Continuous-time Neural Networks to Learn Dependence Structures

Diffusion Denoising as a Certified Defense Against Clean-Label Poisoning Attacks