6 papers
One Modality to Forget Them All: Enhancing Cross-Modal Unlearning in Vision-Language Models
Sudharshan Balaji, Yili Ren, Guangjing Wang +2
Machine unlearning is widely used to remove hazardous knowledge from large language models. Modern Vision-Language Models (VLMs), however, process both text and visual inputs, rais…
PrivScope: Task-scoped Disclosure Control for Hybrid Agentic Systems
Shafizur Rahman Seeam, Zhengxiong Li, Zhiyuan Yu +3
Hybrid local--cloud agents enrich user requests with context from persistent working state before delegating capability-intensive subtasks to a cloud language model (CLM). While th…
ALDEN: Boosting Private Data Extraction from Retrieval-Augmented Generation Systems via Active Learning and Distribution Estimation
Xingyu Lyu, Jianfeng He, Ning Wang +5
Retrieval-Augmented Generation (RAG) is widely used to augment large language models with external knowledge retrieval to improve reliability and generalization. However, recent st…
BoBa: Boosting Backdoor Detection through Data Distribution Inference in Federated Learning
Zhengyuan Jiang, Xingyu Lyu, Shanghao Shi +5
Federated learning, while being a promising approach for collaborative model training, is susceptible to backdoor attacks due to its decentralized nature. Backdoor attacks have sho…
Buffer is All You Need: Defending Federated Learning against Backdoor Attacks under Non-iids via Buffering
Xingyu Lyu, Ning Wang, Yang Xiao +4
Federated Learning (FL) is a popular paradigm enabling clients to jointly train a global model without sharing raw data. However, FL is known to be vulnerable towards backdoor atta…
Two Heads Are Better than One: Model-Weight and Latent-Space Analysis for Federated Learning on Non-iid Data against Poisoning Attacks
Xingyu Lyu, Ning Wang, Yang Xiao +4
Federated Learning is a popular paradigm that enables remote clients to jointly train a global model without sharing their raw data. However, FL has been shown to be vulnerable tow…