Zobrazeno 1 - 10
of 6 898
pro vyhledávání: '"HUANG, Hua"'
Existing single-image denoising algorithms often struggle to restore details when dealing with complex noisy images. The introduction of near-infrared (NIR) images offers new possibilities for RGB image denoising. However, due to the inconsistency be
Externí odkaz:
http://arxiv.org/abs/2412.16645
Image denoising enhances image quality, serving as a foundational technique across various computational photography applications. The obstacle to clean image acquisition in real scenarios necessitates the development of self-supervised image denoisi
Externí odkaz:
http://arxiv.org/abs/2412.16460
Low-light image enhancement (LLIE) is a fundamental task in computational photography, aiming to improve illumination, reduce noise, and enhance the image quality of low-light images. While recent advancements primarily focus on customizing complex n
Externí odkaz:
http://arxiv.org/abs/2412.16459
Recent advances in Large Language Models (LLMs) have demonstrated promising performance in sequential recommendation tasks, leveraging their superior language understanding capabilities. However, existing LLM-based recommendation approaches predomina
Externí odkaz:
http://arxiv.org/abs/2412.05543
Owing to the diverse scales and varying distributions of sparse matrices arising from practical problems, a multitude of choices are present in the design and implementation of sparse matrix-vector multiplication (SpMV). Researchers have proposed man
Externí odkaz:
http://arxiv.org/abs/2411.10143
Iterative solvers are frequently used in scientific applications and engineering computations. However, the memory-bound Sparse Matrix-Vector (SpMV) kernel computation hinders the efficiency of iterative algorithms. As modern hardware increasingly su
Externí odkaz:
http://arxiv.org/abs/2411.04686
Autor:
Zhao, Zijia, Guo, Longteng, Yue, Tongtian, Hu, Erdong, Shao, Shuai, Yuan, Zehuan, Huang, Hua, Liu, Jing
In this paper, we investigate the task of general conversational image retrieval on open-domain images. The objective is to search for images based on interactive conversations between humans and computers. To advance this task, we curate a dataset c
Externí odkaz:
http://arxiv.org/abs/2410.18715
Fine-tuning Large Language Models (LLMs) has proven effective for a variety of downstream tasks. However, as LLMs grow in size, the memory demands for backpropagation become increasingly prohibitive. Zeroth-order (ZO) optimization methods offer a mem
Externí odkaz:
http://arxiv.org/abs/2410.08989
Effective activation functions introduce non-linear transformations, providing neural networks with stronger fitting capa-bilities, which help them better adapt to real data distributions. Huawei Noah's Lab believes that dynamic activation functions
Externí odkaz:
http://arxiv.org/abs/2409.08283
Autor:
Fornaro, James M.1 (AUTHOR), Huang, Hua-Wei2 (AUTHOR) hwawei7@yahoo.com.tw, Lin, Yi-Hung3 (AUTHOR)
Publikováno v:
Journal of Accounting, Auditing & Finance. Jan2025, Vol. 40 Issue 1, p101-138. 38p.
