Zobrazeno 1 - 10
of 2 355
pro vyhledávání: '"KIM, Hyungjun"'
Autor:
Kim, Taesu, Lee, Jongho, Ahn, Daehyun, Kim, Sarang, Choi, Jiwoong, Kim, Minkyu, Kim, Hyungjun
We introduce QUICK, a group of novel optimized CUDA kernels for the efficient inference of quantized Large Language Models (LLMs). QUICK addresses the shared memory bank-conflict problem of state-of-the-art mixed precision matrix multiplication kerne
Externí odkaz:
http://arxiv.org/abs/2402.10076
Large language models (LLMs) have proven to be highly effective across various natural language processing tasks. However, their large number of parameters poses significant challenges for practical deployment. Pruning, a technique aimed at reducing
Externí odkaz:
http://arxiv.org/abs/2402.09025
Autor:
Choi, Jiwoong, Kim, Minkyu, Ahn, Daehyun, Kim, Taesu, Kim, Yulhwa, Jo, Dongwon, Jeon, Hyesung, Kim, Jae-Joon, Kim, Hyungjun
The emergence of diffusion models has greatly broadened the scope of high-fidelity image synthesis, resulting in notable advancements in both practical implementation and academic research. With the active adoption of the model in various real-world
Externí odkaz:
http://arxiv.org/abs/2307.01193
The diffusion model has gained popularity in vision applications due to its remarkable generative performance and versatility. However, high storage and computation demands, resulting from the model size and iterative generation, hinder its use on mo
Externí odkaz:
http://arxiv.org/abs/2306.02316
Large language models (LLMs) with hundreds of billions of parameters require powerful server-grade GPUs for inference, limiting their practical deployment. To address this challenge, we introduce the outlier-aware weight quantization (OWQ) method, wh
Externí odkaz:
http://arxiv.org/abs/2306.02272
Binary Neural Networks (BNNs) have emerged as a promising solution for reducing the memory footprint and compute costs of deep neural networks, but they suffer from quality degradation due to the lack of freedom as activations and weights are constra
Externí odkaz:
http://arxiv.org/abs/2204.07439
Autor:
Khaliq, Nisar Ul, Lee, Juyeon, Kim, Yejin, Kim, Joohyeon, Kim, Taeho, Yu, Sohyeon, Seo, Dongseong, Sung, Daekyung, Kim, Hyungjun
Publikováno v:
In BBA - General Subjects November 2024 1868(11)
Publikováno v:
In Chemical Engineering Journal 1 October 2024 497
Publikováno v:
In Applied Surface Science 15 February 2025 682