Expected Harm: Rethinking Safety Evaluation of (Mis)Aligned LLMs
Paper
•
2602.01600
•
Published
•
18
Jiawei
jwliao1209
·
AI & ML interests
Diffusion Models, Inference-time Optimization, Protective Generative AI
Recent Activity
upvoted
a
paper
about 19 hours ago
Expected Harm: Rethinking Safety Evaluation of (Mis)Aligned LLMs
authored
a paper
29 days ago
M-ErasureBench: A Comprehensive Multimodal Evaluation Benchmark for Concept Erasure in Diffusion Models
submitted
a paper
29 days ago
M-ErasureBench: A Comprehensive Multimodal Evaluation Benchmark for Concept Erasure in Diffusion Models
Organizations
upvoted
a
paper
about 19 hours ago