Search Results - Seong, Haebin
- Showing 1 - 7 results of 7
-
1
BiasJailbreak:Analyzing Ethical Biases and Jailbreak Vulnerabilities in Large Language Models Autoría: Lee, Isack, Seong, Haebin
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
2
The Last Harness You'll Ever Build Autoría: Seong, Haebin, Yin, Li, Zhang, Haoran, Shi, Zhan
Published 2026Fuente: arXivTipo de material: PreprintAcceso al recurso -
3
SafeRoute: Adaptive Model Selection for Efficient and Accurate Safety Guardrails in Large Language Models Autoría: Lee, Seanie, Lee, Dong Bok, Wagner, Dominik, Kang, Minki, Seong, Haebin, Bocklet, Tobias, Lee, Juho, Hwang, Sung Ju
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso -
4
FedSVD: Adaptive Orthogonalization for Private Federated Learning with LoRA Autoría: Lee, Seanie, Park, Sangwoo, Lee, Dong Bok, Wagner, Dominik, Seong, Haebin, Bocklet, Tobias, Lee, Juho, Hwang, Sung Ju
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso -
5
HarmAug: Effective Data Augmentation for Knowledge Distillation of Safety Guard Models Autoría: Lee, Seanie, Seong, Haebin, Lee, Dong Bok, Kang, Minki, Chen, Xiaoyin, Wagner, Dominik, Bengio, Yoshua, Lee, Juho, Hwang, Sung Ju
Published 2024Fuente: arXivTipo de material: PreprintAcceso al recurso -
6
D2E: Scaling Vision-Action Pretraining on Desktop Data for Transfer to Embodied AI Autoría: Choi, Suhwan, Jung, Jaeyoon, Seong, Haebin, Kim, Minchan, Kim, Minyeong, Cho, Yongjun, Kim, Yoonshik, Park, Yubeen, Yu, Youngjae, Lee, Yunsung
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso -
7
CostNav: A Navigation Benchmark for Real-World Economic-Cost Evaluation of Physical AI Agents Autoría: Seong, Haebin, Kim, Sungmin, Cho, Yongjun, Joe, Myunchul, Kim, Geunwoo, Park, Yubeen, Kim, Sunhoo, Kim, Yoonshik, Choi, Suhwan, Jung, Jaeyoon, Youn, Jiyong, Kwak, Jinmyung, Ahn, Sunghee, Lee, Jaemin, Do, Younggil, Yi, Seungyeop, Cheong, Woojin, Oh, Minhyeok, Kim, Minchan, Kang, Seongjae, Seong, Samwoo, Yu, Youngjae, Lee, Yunsung
Published 2025Fuente: arXivTipo de material: PreprintAcceso al recurso