Towards Mitigating Deceptive Safety Alignment in Large Reasoning Models
Paper • 2609.36254 • Published • 1
Xiangyu Zhou
XiangyuZhou
AI & ML interests
None yet
Recent Activity
authored a paper 4 days ago
Towards Mitigating Deceptive Safety Alignment in Large Reasoning Models authored a paper 4 days ago
SpatialCORE: Confidence-Aware Grounded Spatial Reasoning in Large Vision--Language Models authored a paper 4 days ago
Attention Smoothing Is All You Need For UnlearningOrganizations
None yet
authored 6
papers 4 days ago
SpatialCORE: Confidence-Aware Grounded Spatial Reasoning in Large Vision--Language Models
Paper • 2609.38716 • Published • 4
Attention Smoothing Is All You Need For Unlearning
Paper • 2603.01285 • Published
WalkGPT: Grounded Vision-Language Conversation with Depth-Aware Segmentation for Pedestrian Navigation
Paper • 2603.10703 • Published
Learning to Poison Large Language Models for Downstream Manipulation
Paper • 2402.13459 • Published
Not All Tokens Are Meant to Be Forgotten
Paper • 2506.03142 • Published
upvoted a paper 4 days ago
liked a
dataset about 1 month ago