Showing 2026Show all
2 papers · 1 filter
cs.AI2026
On the Steering Dimensionality of Refusal in Language Models
Han Wang, Erik Miehling, Dennis Wei +2
Existing activation steering methods often assume that a high-level concept can be mediated by a single steering direction. To support this, two complementary interventions should…
cs.CV2026
Taming Noise-Induced Prototype Degradation for Privacy-Preserving Personalized Federated Fine-Tuning
Yuhua Wang, Qinnan Zhang, Xiaodong Li +6
Prototype-based Personalized Federated Learning (ProtoPFL) enables efficient multi-domain adaptation by communicating compact class prototypes, but directly sharing them poses priv…