Zobrazeno 1 - 10
of 12 194
pro vyhledávání: '"Kim BO"'
Photon-magnon coupling, where electromagnetic waves interact with spin waves, and negative refraction, which bends the direction of electromagnetic waves unnaturally, constitute critical foundations and advancements in the realms of optics, spintroni
Externí odkaz:
http://arxiv.org/abs/2406.18858
Latent Diffusion Models (LDMs) have emerged as powerful generative models, known for delivering remarkable results under constrained computational resources. However, deploying LDMs on resource-limited devices remains a complex issue, presenting chal
Externí odkaz:
http://arxiv.org/abs/2404.11936
Autor:
Castells, Thibault, Song, Hyoung-Kyu, Piao, Tairen, Choi, Shinkook, Kim, Bo-Kyeong, Yim, Hanyoung, Lee, Changgwun, Kim, Jae Gon, Kim, Tae-Ho
The intensive computational burden of Stable Diffusion (SD) for text-to-image generation poses a significant hurdle for its practical application. To tackle this challenge, recent research focuses on methods to reduce sampling steps, such as Latent C
Externí odkaz:
http://arxiv.org/abs/2404.11925
Autor:
Kim, Bo-Kyeong, Kim, Geonmin, Kim, Tae-Ho, Castells, Thibault, Choi, Shinkook, Shin, Junho, Song, Hyoung-Kyu
Structured pruning of modern large language models (LLMs) has emerged as a way of decreasing their high computational needs. Width pruning reduces the size of projection weight matrices (e.g., by removing attention heads) while maintaining the number
Externí odkaz:
http://arxiv.org/abs/2402.02834
Text-to-image (T2I) generation with Stable Diffusion models (SDMs) involves high computing demands due to billion-scale parameters. To enhance efficiency, recent studies have reduced sampling steps and applied network quantization while retaining the
Externí odkaz:
http://arxiv.org/abs/2305.15798
Autor:
Kim, Bo-Kyeong, Kang, Jaemin, Seo, Daeun, Park, Hancheol, Choi, Shinkook, Song, Hyoung-Kyu, Kim, Hyungshin, Lim, Sungsu
Virtual humans have gained considerable attention in numerous industries, e.g., entertainment and e-commerce. As a core technology, synthesizing photorealistic face frames from target speech and facial identity has been actively studied with generati
Externí odkaz:
http://arxiv.org/abs/2304.00471
Publikováno v:
PLoS ONE. 11/26/2024, Vol. 19 Issue 11, p1-19. 19p.
Autor:
Kim, Bo-Guen1 (AUTHOR) kbg1q2w3e@gmail.com, Yoon, Sanghyuk2 (AUTHOR) whysh313@gmail.com, Lee, Sun Yeop2 (AUTHOR) kimeg9160@gmail.com, Kim, Eun Gyo2 (AUTHOR) jokim8505@gmail.com, Kim, Jung Oh2 (AUTHOR), Kim, Jong Seung3,4 (AUTHOR) kjsjdk@gmail.com, Lee, Hyun5 (AUTHOR) kjsjdk@gmail.com
Publikováno v:
Journal of Clinical Medicine. Oct2024, Vol. 13 Issue 20, p6080. 10p.
Pruning effectively compresses overparameterized models. Despite the success of pruning methods for discriminative models, applying them for generative models has been relatively rarely approached. This study conducts structured pruning on U-Net gene
Externí odkaz:
http://arxiv.org/abs/2206.14658
Publikováno v:
In Applied Acoustics 5 November 2024 225