Zobrazeno 1 - 10
of 27 846
pro vyhledávání: '"To Ngai"'
Speech enhancement is crucial in human-computer interaction, especially for ubiquitous devices. Ultrasound-based speech enhancement has emerged as an attractive choice because of its superior ubiquity and performance. However, inevitable interference
Externí odkaz:
http://arxiv.org/abs/2410.22076
The metaverse, emerging as a revolutionary platform for social and economic activities, provides various virtual services while posing security and privacy challenges. Wearable devices serve as bridges between the real world and the metaverse. To pro
Externí odkaz:
http://arxiv.org/abs/2410.21675
Recently, prompt learning has emerged as the state-of-the-art (SOTA) for fair text-to-image (T2I) generation. Specifically, this approach leverages readily available reference images to learn inclusive prompts for each target Sensitive Attribute (tSA
Externí odkaz:
http://arxiv.org/abs/2410.18615
Autor:
Yang, Zebin, Chen, Renze, Wu, Taiqiang, Wong, Ngai, Liang, Yun, Wang, Runsheng, Huang, Ru, Li, Meng
In this paper, we propose MCUBERT to enable language models like BERT on tiny microcontroller units (MCUs) through network and scheduling co-optimization. We observe the embedding table contributes to the major storage bottleneck for tiny BERT models
Externí odkaz:
http://arxiv.org/abs/2410.17957
Autor:
Chen, Handi, Deng, Weipeng, Yang, Shuo, Xu, Jinfeng, Jiang, Zhihan, Ngai, Edith C. H., Liu, Jiangchuan, Liu, Xue
Edge Intelligence (EI) has been instrumental in delivering real-time, localized services by leveraging the computational capabilities of edge networks. The integration of Large Language Models (LLMs) empowers EI to evolve into the next stage: Edge Ge
Externí odkaz:
http://arxiv.org/abs/2410.18125
Autor:
Xiong, Jing, Shen, Jianghan, Ye, Fanghua, Tao, Chaofan, Wan, Zhongwei, Lu, Jianqiao, Wu, Xun, Zheng, Chuanyang, Guo, Zhijiang, Kong, Lingpeng, Wong, Ngai
Deploying large language models (LLMs) is challenging due to their high memory and computational demands, especially during long-context inference. While key-value (KV) caching accelerates inference by reusing previously computed keys and values, it
Externí odkaz:
http://arxiv.org/abs/2410.03090
Autor:
Li, Zixuan, Xiong, Jing, Ye, Fanghua, Zheng, Chuanyang, Wu, Xun, Lu, Jianqiao, Wan, Zhongwei, Liang, Xiaodan, Li, Chengming, Sun, Zhenan, Kong, Lingpeng, Wong, Ngai
We present UncertaintyRAG, a novel approach for long-context Retrieval-Augmented Generation (RAG) that utilizes Signal-to-Noise Ratio (SNR)-based span uncertainty to estimate similarity between text chunks. This span uncertainty enhances model calibr
Externí odkaz:
http://arxiv.org/abs/2410.02719
Autor:
Dodds, Tomás, Vandendaele, Astrid, Simon, Felix M., Helberger, Natali, Resendez, Valeria, Yeung, Wang Ngai
The effective adoption of responsible AI practices in journalism requires a concerted effort to bridge different perspectives, including technological, editorial, journalistic, and managerial. Among the many challenges that could impact information s
Externí odkaz:
http://arxiv.org/abs/2410.01138
Autor:
Li, Siheng, Yang, Cheng, Wu, Taiqiang, Shi, Chufan, Zhang, Yuji, Zhu, Xinyu, Cheng, Zesen, Cai, Deng, Yu, Mo, Liu, Lemao, Zhou, Jie, Yang, Yujiu, Wong, Ngai, Wu, Xixin, Lam, Wai
Honesty is a fundamental principle for aligning large language models (LLMs) with human values, requiring these models to recognize what they know and don't know and be able to faithfully express their knowledge. Despite promising, current LLMs still
Externí odkaz:
http://arxiv.org/abs/2409.18786
Phone scams pose a significant threat to individuals and communities, causing substantial financial losses and emotional distress. Despite ongoing efforts to combat these scams, scammers continue to adapt and refine their tactics, making it imperativ
Externí odkaz:
http://arxiv.org/abs/2409.11643