Zobrazeno 1 - 10
of 13 230
pro vyhledávání: '"An, Sanghyun"'
Recent advancements in Large Language Models (LLMs) have been remarkable, with new models consistently surpassing their predecessors. These advancements are underpinned by extensive research on various training mechanisms. Among these, Preference Opt
Externí odkaz:
http://arxiv.org/abs/2412.07812
Autor:
Coalson, Zachary, Woo, Jeonghyun, Chen, Shiyang, Sun, Yu, Yang, Lishan, Nair, Prashant, Fang, Bo, Hong, Sanghyun
We introduce a new class of attacks on commercial-scale (human-aligned) language models that induce jailbreaking through targeted bitwise corruptions in model parameters. Our adversary can jailbreak billion-parameter language models with fewer than 2
Externí odkaz:
http://arxiv.org/abs/2412.07192
Autor:
Piao, Shengmin, Park, Sanghyun
Large Language Models exhibit impressive reasoning capabilities across diverse tasks, motivating efforts to distill these capabilities into smaller models through generated reasoning data. However, direct training on such synthesized reasoning data m
Externí odkaz:
http://arxiv.org/abs/2412.08024
Autor:
Ahn, Donghoon, Kang, Jiwon, Lee, Sanghyun, Min, Jaewon, Kim, Minjae, Jang, Wooseok, Cho, Hyoungwon, Paul, Sayak, Kim, SeonHwa, Cha, Eunju, Jin, Kyong Hwan, Kim, Seungryong
Diffusion models excel in generating high-quality images. However, current diffusion models struggle to produce reliable images without guidance methods, such as classifier-free guidance (CFG). Are guidance methods truly necessary? Observing that noi
Externí odkaz:
http://arxiv.org/abs/2412.03895
We propose a Multifaceted Resilient Network(MRNet), a novel architecture developed for medical image-to-image translation that outperforms state-of-the-art methods in MRI-to-CT and MRI-to-MRI conversion. MRNet leverages the Segment Anything Model (SA
Externí odkaz:
http://arxiv.org/abs/2412.03039
Fine-tuning text-to-image diffusion models is widely used for personalization and adaptation for new domains. In this paper, we identify a critical vulnerability of fine-tuning: safety alignment methods designed to filter harmful content (e.g., nudit
Externí odkaz:
http://arxiv.org/abs/2412.00357
In this paper, we develop monolithic limiting techniques for enforcing nonlinear stability constraints in enriched Galerkin (EG) discretizations of nonlinear scalar hyperbolic equations. To achieve local mass conservation and gain control over the ce
Externí odkaz:
http://arxiv.org/abs/2411.19160
Autor:
Zhao, Xuandong, Gunn, Sam, Christ, Miranda, Fairoze, Jaiden, Fabrega, Andres, Carlini, Nicholas, Garg, Sanjam, Hong, Sanghyun, Nasr, Milad, Tramer, Florian, Jha, Somesh, Li, Lei, Wang, Yu-Xiang, Song, Dawn
As the outputs of generative AI (GenAI) techniques improve in quality, it becomes increasingly challenging to distinguish them from human-created content. Watermarking schemes are a promising approach to address the problem of distinguishing between
Externí odkaz:
http://arxiv.org/abs/2411.18479
Multi-modal sensor fusion in bird's-eye-view (BEV) representation has become the leading approach in 3D object detection. However, existing methods often rely on depth estimators or transformer encoders for view transformation, incurring substantial
Externí odkaz:
http://arxiv.org/abs/2411.10715
Autor:
Byun, Sanghyun, Shah, Kayvan, Gang, Ayushi, Apton, Christopher, Song, Jacob, Chung, Woo Seong
Many state-of-the-art computer vision architectures leverage U-Net for its adaptability and efficient feature extraction. However, the multi-resolution convolutional design often leads to significant computational demands, limiting deployment on edge
Externí odkaz:
http://arxiv.org/abs/2411.09838