Zobrazeno 1 - 10
of 2 410
pro vyhledávání: '"Min, Rui"'
Reward models are critical in techniques like Reinforcement Learning from Human Feedback (RLHF) and Inference Scaling Laws, where they guide language model alignment and select optimal responses. Despite their importance, existing reward model benchm
Externí odkaz:
http://arxiv.org/abs/2410.16184
Backdoor attacks pose a significant threat to Deep Neural Networks (DNNs) as they allow attackers to manipulate model predictions with backdoor triggers. To address these security vulnerabilities, various backdoor purification methods have been propo
Externí odkaz:
http://arxiv.org/abs/2410.09838
Autor:
Besta, Maciej, Gerstenberger, Robert, Iff, Patrick, Sonawane, Pournima, Luna, Juan Gómez, Kanakagiri, Raghavendra, Min, Rui, Mutlu, Onur, Hoefler, Torsten, Appuswamy, Raja, Mahony, Aidan O
Knowledge graphs (KGs) have achieved significant attention in recent years, particularly in the area of the Semantic Web as well as gaining popularity in other application domains such as data mining and search engines. Simultaneously, there has been
Externí odkaz:
http://arxiv.org/abs/2408.12173
The ethical need to protect AI-generated content has been a significant concern in recent years. While existing watermarking strategies have demonstrated success in detecting synthetic content (detection), there has been limited exploration in identi
Externí odkaz:
http://arxiv.org/abs/2403.10893
It has been widely observed that deep neural networks (DNN) are vulnerable to backdoor attacks where attackers could manipulate the model behavior maliciously by tampering with a small set of training samples. Although a line of defense methods is pr
Externí odkaz:
http://arxiv.org/abs/2310.01875
Autor:
Xiao-Long Zhang, Peng-Cheng Yu, Shu-Ping Sun, Lei Shi, Peng-Peng Yang, Zhi-Zheng Wu, Li-Ping Chi, Ya-Rong Zheng, Min-Rui Gao
Publikováno v:
Nature Communications, Vol 15, Iss 1, Pp 1-11 (2024)
Abstract Seawater electrolysis using renewable electricity offers an attractive route to sustainable hydrogen production, but the sluggish electrode kinetics and poor durability are two major challenges. We report a molybdenum nitride (Mo2N) catalyst
Externí odkaz:
https://doaj.org/article/6f69a32550de4ab9b078a4091d8b8860
Autor:
Yu Shen, Xiao-Long Zhang, Ming-Rong Qu, Jie Ma, Sheng Zhu, Yu-Lin Min, Min-Rui Gao, Shu-Hong Yu
Publikováno v:
Nature Communications, Vol 15, Iss 1, Pp 1-12 (2024)
Abstract Simultaneously improving the activity and stability of catalysts for anodic oxygen evolution reaction (OER) in proton exchange membrane water electrolysis (PEMWE) remains a notable challenge. Here, we report a chromium-doped ruthenium dioxid
Externí odkaz:
https://doaj.org/article/f8a3457d700c4d0fa587b1f04767d258
Autor:
Cindy Q. Tang, Min-Rui Du, Huan-Chong Wang, You-Cai Shi, Jia-Le Zeng, Shu-Li Xiao, Peng-Bin Han, Jian-Ran Wen, Shi-Qian Yao, Ming-Chun Peng, Chong-Yun Wang, Yong-Ping Li, Jordi López-Pujol
Publikováno v:
Plant Diversity, Vol 46, Iss 5, Pp 648-660 (2024)
Relict subtropical coniferous forests in China face severe fragmentation, resulting in declining populations, and some are under significant threat from invasive alien species. Despite the crucial importance of understanding forest dynamics, knowledg
Externí odkaz:
https://doaj.org/article/9271263de62a4577a98a5eb35bc27dbc
Autor:
Wang, Yuhang, Shi, Huafeng, Min, Rui, Wu, Ruijia, Liang, Siyuan, Wu, Yichao, Liang, Ding, Liu, Aishan
Extensive evidence has demonstrated that deep neural networks (DNNs) are vulnerable to backdoor attacks, which motivates the development of backdoor attacks detection. Most detection methods are designed to verify whether a model is infected with pre
Externí odkaz:
http://arxiv.org/abs/2209.05244
Autor:
Jing-Wen DuanMu, Zhi-Zheng Wu, Fei-Yue Gao, Peng-Peng Yang, Zhuang-Zhuang Niu, Yu-Cai Zhang, Li-Ping Chi, Min-Rui Gao
Publikováno v:
Precision Chemistry, Vol 2, Iss 4, Pp 151-160 (2024)
Externí odkaz:
https://doaj.org/article/5819f1ecc38d4e6186e65bebd2cf614e