Ling Liu

papers

182

total citations

papers (11)

Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation

ICLR 2025arXiv

citations

TACO: Benchmarking Generalizable Bimanual Tool-ACtion-Object Understanding

CVPR 2024arXiv

citations

Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning Attack

ICML 2025

citations

Lockdown: Backdoor Defense for Federated Learning with Isolated Subspace Training

NEURIPS 2023

citations

ScaleFL: Resource-Adaptive Federated Learning With Heterogeneous Clients

CVPR 2023

citations

Resource-Efficient Transformer Pruning for Finetuning of Large Models

CVPR 2024

citations

Boosting Ensemble Accuracy by Revisiting Ensemble Diversity Metrics

CVPR 2021

citations

Ling Liu

papers (11)

Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation

TACO: Benchmarking Generalizable Bimanual Tool-ACtion-Object Understanding

Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning Attack

STDLens: Model Hijacking-Resilient Federated Learning for Object Detection

Personalized Privacy Protection Mask Against Unauthorized Facial Recognition

Adversarial Attention Perturbations for Large Object Detection Transformers

4DSegStreamer: Streaming 4D Panoptic Segmentation via Dual Threads

Lockdown: Backdoor Defense for Federated Learning with Isolated Subspace Training

ScaleFL: Resource-Adaptive Federated Learning With Heterogeneous Clients

Resource-Efficient Transformer Pruning for Finetuning of Large Models

Boosting Ensemble Accuracy by Revisiting Ensemble Diversity Metrics

papers (11)

Booster: Tackling Harmful Fine-tuning for Large Language Models via Attenuating Harmful Perturbation

TACO: Benchmarking Generalizable Bimanual Tool-ACtion-Object Understanding

Antidote: Post-fine-tuning Safety Alignment for Large Language Models against Harmful Fine-tuning Attack

STDLens: Model Hijacking-Resilient Federated Learning for Object Detection

Personalized Privacy Protection Mask Against Unauthorized Facial Recognition

Adversarial Attention Perturbations for Large Object Detection Transformers

4DSegStreamer: Streaming 4D Panoptic Segmentation via Dual Threads

Lockdown: Backdoor Defense for Federated Learning with Isolated Subspace Training

ScaleFL: Resource-Adaptive Federated Learning With Heterogeneous Clients

Resource-Efficient Transformer Pruning for Finetuning of Large Models

Boosting Ensemble Accuracy by Revisiting Ensemble Diversity Metrics