Picture for Jiancheng Liu

Jiancheng Liu

Towards Universal Mesh Movement Networks

Add code
Jul 02, 2024
Viaarxiv icon

Label Smoothing Improves Machine Unlearning

Add code
Jun 11, 2024
Viaarxiv icon

Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models

Add code
May 24, 2024
Viaarxiv icon

SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning

Add code
Apr 28, 2024
Figure 1 for SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning
Figure 2 for SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning
Figure 3 for SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning
Figure 4 for SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning
Viaarxiv icon

Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning

Add code
Mar 12, 2024
Figure 1 for Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning
Figure 2 for Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning
Figure 3 for Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning
Figure 4 for Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning
Viaarxiv icon

UnlearnCanvas: A Stylized Image Dataset to Benchmark Machine Unlearning for Diffusion Models

Add code
Feb 26, 2024
Viaarxiv icon

SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation

Add code
Oct 19, 2023
Figure 1 for SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation
Figure 2 for SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation
Figure 3 for SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation
Figure 4 for SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation
Viaarxiv icon

To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images For Now

Add code
Oct 18, 2023
Figure 1 for To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images  For Now
Figure 2 for To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images  For Now
Figure 3 for To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images  For Now
Figure 4 for To Generate or Not? Safety-Driven Unlearned Diffusion Models Are Still Easy To Generate Unsafe Images  For Now
Viaarxiv icon

Selectivity Drives Productivity: Efficient Dataset Pruning for Enhanced Transfer Learning

Add code
Oct 16, 2023
Viaarxiv icon

DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training

Add code
Oct 03, 2023
Figure 1 for DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training
Figure 2 for DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training
Figure 3 for DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training
Figure 4 for DeepZero: Scaling up Zeroth-Order Optimization for Deep Model Training
Viaarxiv icon