5 papers
SafeBranch: Branch-Pair Safety Alignment for Embodied Agents
Hyunse Lee, Jiwoo Jeong, Haneul Lee +3
Vision-language-model-based embodied agents can complete instructed tasks but often violate safety constraints in the process, a problem recently framed as interactive safety. Trai…
SlotGCG: Exploiting the Positional Vulnerability in LLMs for Jailbreak Attacks
Seungwon Jeong, Jiwoo Jeong, Hyeonjin Kim +2
As large language models (LLMs) are widely deployed, identifying their vulnerability through jailbreak attacks becomes increasingly critical. Optimization-based attacks like Greedy…
Machine Unlearning for Masked Diffusion Language Models
Georu Lee, Seungwon Jeong, Hoki Kim +2
Recent masked diffusion language models (MDLMs), such as LLaDA and Dream, have achieved performance comparable to autoregressive large language models. Unlike autoregressive models…
SeMoBridge: Semantic Modality Bridge for Efficient Few-Shot Adaptation of CLIP
Christoph Timmermann, Hyunse Lee, Woojin Lee
While Contrastive Language-Image Pretraining (CLIP) excels at zero-shot tasks by aligning image and text embeddings, its performance in few-shot classification is hindered by a cri…
CaddieSet: A Golf Swing Dataset with Human Joint Features and Ball Information
Seunghyeon Jung, Seoyoung Hong, Jiwoo Jeong +4
Recent advances in deep learning have led to more studies to enhance golfers' shot precision. However, these existing studies have not quantitatively established the relationship b…