3 papers
cs.AI2026
DeShortcut-Align: Decoupling Spurious Shortcuts for Robust Safety Alignment in Large Reasoning Models
Qirui Liu, Yichen Sun, Yan Wang +4
Safety alignment of large reasoning models (LRMs) via supervised fine-tuning (SFT) and reinforcement learning (RL) often yields near-perfect safety scores, yet this apparent succes…
cs.IR2025
FedAU2: Attribute Unlearning for User-Level Federated Recommender Systems with Adaptive and Robust Adversarial Training
Yuyuan Li, Junjie Fang, Fengyuan Yu +7
Federated Recommender Systems (FedRecs) leverage federated learning to protect user privacy by retaining data locally. However, user embeddings in FedRecs often encode sensitive at…
cs.AI2025
TOFA: Training-Free One-Shot Federated Adaptation for Vision-Language Models
Li Zhang, Zhongxuan Han, XiaoHua Feng +5
Efficient and lightweight adaptation of pre-trained Vision-Language Models (VLMs) to downstream tasks through collaborative interactions between local clients and a central server…