mz-kim

mz-kim

·

AI & ML interests

None yet

Recent Activity

upvoted a paper 1 day ago

Refuse without Refusal: A Structural Analysis of Safety-Tuning Responses for Reducing False Refusals in Language Models

upvoted a paper 1 day ago

Knowing What Not to Answer: Selective Non-Compliance in Vision-Language Models

new activity 11 days ago

huggingface/HuggingDiscussions:[FEEDBACK] Daily Papers

View all activity

Organizations

Papers 3

arxiv:2609.04714

arxiv:2609.04720

arxiv:2506.00421

models 0

None public yet

datasets 1

mz-kim/KoNA

Viewer • Updated 12 days ago • 3.1k • 99