Zobrazeno 1 - 10
of 28 027
pro vyhledávání: '"Wu, Di."'
Autor:
Song, Xingchen, Liang, Chengdong, Zhang, Binbin, Zhang, Pengshen, Wang, ZiYu, Ma, Youcheng, Xu, Menglong, Wang, Lin, Wu, Di, Pan, Fuping, Zhou, Dinghao, Peng, Zhendong
Large Automatic Speech Recognition (ASR) models demand a vast number of parameters, copious amounts of data, and significant computational resources during the training process. However, such models can merely be deployed on high-compute cloud platfo
Externí odkaz:
http://arxiv.org/abs/2412.15622
Channel knowledge map (CKM) is a promising technique that enables environment-aware wireless networks by utilizing location-specific channel prior information to improve communication and sensing performance. A fundamental problem for CKM constructio
Externí odkaz:
http://arxiv.org/abs/2412.14812
Autor:
Wu, Kun, Hou, Chengkai, Liu, Jiaming, Che, Zhengping, Ju, Xiaozhu, Yang, Zhuqin, Li, Meng, Zhao, Yinuo, Xu, Zhiyuan, Yang, Guang, Zhao, Zhen, Li, Guangyu, Jin, Zhao, Wang, Lecheng, Mao, Jilei, Wang, Xinhua, Fan, Shichao, Liu, Ning, Ren, Pei, Zhang, Qiang, Lyu, Yaoxu, Liu, Mengzhen, He, Jingyang, Luo, Yulin, Gao, Zeyu, Li, Chenxuan, Gu, Chenyang, Fu, Yankai, Wu, Di, Wang, Xingyu, Chen, Sixiang, Wang, Zhenyu, An, Pengju, Qian, Siyuan, Zhang, Shanghang, Tang, Jian
Developing robust and general-purpose robotic manipulation policies is a key goal in the field of robotics. To achieve effective generalization, it is essential to construct comprehensive datasets that encompass a large number of demonstration trajec
Externí odkaz:
http://arxiv.org/abs/2412.13877
Although large language models (LLMs) achieve effective safety alignment at the time of release, they still face various safety challenges. A key issue is that fine-tuning often compromises the safety alignment of LLMs. To address this issue, we prop
Externí odkaz:
http://arxiv.org/abs/2412.11041
Autor:
Song, Xingchen, Xing, Mengtao, Ma, Changwei, Li, Shengqiang, Wu, Di, Zhang, Binbin, Pan, Fuping, Zhou, Dinghao, Zhang, Yuekai, Lei, Shun, Peng, Zhendong, Wu, Zhiyong
It is well known that LLM-based systems are data-hungry. Recent LLM-based TTS works typically employ complex data processing pipelines to obtain high-quality training data. These sophisticated pipelines require excellent models at each stage (e.g., s
Externí odkaz:
http://arxiv.org/abs/2412.08237
Autor:
Zhao, Wan-Qian, Guo, Zhan-Yong, Tian, Zeng-Yuan, Su, Tong-Fu, Cao, Gang-Qiang, Qi, Zi-Xin, Qin, Tian-Cang, Zhou, Wei, Yang, Jin-Yu, Chen, Ming-Jie, Zhang, Xin-Ge, Zhou, Chun-Yan, Zhu, Chuan-Jia, Tang, Meng-Fei, Wu, Di, Song, Mei-Rong, Guo, Yu-Qi, Qiu, Li-You, Liang, Fei, Li, Mei-Jun, Geng, Jun-Hui, Zhao, Li-Juan, Zhang, Shu-Jie
High quality ancient DNA (aDNA) is essential for molecular paleontology. Due to DNA degradation and contamination by environmental DNA (eDNA), current research is limited to fossils less than 1 million years old. The study successfully extracted DNA
Externí odkaz:
http://arxiv.org/abs/2412.06521
Autor:
Qu, Yixiang, Dai, Yifan, Yu, Shilin, Tanikella, Pradham, Schrank, Travis, Hackman, Trevor, Li, Didong, Wu, Di
Large Language Models (LLMs) have shown impressive capabilities in natural language processing, yet their use in sensitive domains like healthcare, particularly with Electronic Health Records (EHR), faces significant challenges due to privacy concern
Externí odkaz:
http://arxiv.org/abs/2412.02868
Autor:
Zhang, Xianzhi, Zhou, Yipeng, Hu, Miao, Wu, Di, Liao, Pengshan, Guizani, Mohsen, Sheng, Michael
To mitigate the rising concern about privacy leakage, the federated recommender (FR) paradigm emerges, in which decentralized clients co-train the recommendation model without exposing their raw user-item rating data. The differentially private feder
Externí odkaz:
http://arxiv.org/abs/2412.02934
In the era of big data, managing evolving graph data poses substantial challenges due to storage costs and privacy issues. Training graph neural networks (GNNs) on such evolving data usually causes catastrophic forgetting, impairing performance on ea
Externí odkaz:
http://arxiv.org/abs/2411.18919
Transformer-based multi-object tracking (MOT) methods have captured the attention of many researchers in recent years. However, these models often suffer from slow inference speeds due to their structure or other issues. To address this problem, we r
Externí odkaz:
http://arxiv.org/abs/2411.15811