Zobrazeno 1 - 10
of 64 511
pro vyhledávání: '"Liu., Jing"'
Lately, the practice of utilizing task-specific fine-tuning has been implemented to improve the performance of large language models (LLM) in subsequent tasks. Through the integration of diverse LLMs, the overall competency of LLMs is significantly b
Externí odkaz:
http://arxiv.org/abs/2412.15283
Community structures are critical for understanding the mesoscopic organization of networks, bridging local and global patterns. While methods such as DeepWalk and node2vec capture local positional information through random walks, they fail to prese
Externí odkaz:
http://arxiv.org/abs/2412.12933
In real-world applications, spectral Graph Neural Networks (GNNs) are powerful tools for processing diverse types of graphs. However, a single GNN often struggles to handle different graph types-such as homogeneous and heterogeneous graphs-simultaneo
Externí odkaz:
http://arxiv.org/abs/2412.12483
Autor:
Shen, Xuan, Song, Zhao, Zhou, Yufa, Chen, Bo, Liu, Jing, Zhang, Ruiyi, Rossi, Ryan A., Tan, Hao, Yu, Tong, Chen, Xiang, Zhou, Yufan, Sun, Tong, Zhao, Pu, Wang, Yanzhi, Gu, Jiuxiang
Transformers have emerged as the leading architecture in deep learning, proving to be versatile and highly effective across diverse domains beyond language and image processing. However, their impressive performance often incurs high computational co
Externí odkaz:
http://arxiv.org/abs/2412.12441
Due to the sensitivity of data, Federated Learning (FL) is employed to enable distributed machine learning while safeguarding data privacy and accommodating the requirements of various devices. However, in the context of semi-decentralized FL, client
Externí odkaz:
http://arxiv.org/abs/2412.11448
Let $\mu_{\{M_n\},\{D_n\}}$ be a Moran measure on $\mathbb{R}^2$ generated by a sequence of expanding matrices $\{M_n\}\subset GL(2, \mathbb{Z})$ and a sequence of integer digit sets $\{D_n\}$ where $D_n=\left\{\begin{pmatrix} 0 \\ 0 \end{pmatrix},\b
Externí odkaz:
http://arxiv.org/abs/2412.11200
Autor:
Jiang, Yudong, Xu, Baohan, Yang, Siqian, Yin, Mingyu, Liu, Jing, Xu, Chao, Wang, Siqi, Wu, Yidi, Zhu, Bingwen, Zhang, Xinwen, Zheng, Xingyu, Xu, Jixuan, Zhang, Yue, Hou, Jinlong, Sun, Huyang
Animation has gained significant interest in the recent film and TV industry. Despite the success of advanced video generation models like Sora, Kling, and CogVideoX in generating natural videos, they lack the same effectiveness in handling animation
Externí odkaz:
http://arxiv.org/abs/2412.10255
Autor:
Liu, Jing, Fourtassi, Abdellah
LLMs can generate human-like dialogues, yet their ability to simulate early child-adult interactions remains largely unexplored. In this paper, we examined how effectively LLMs can capture the distinctive features of child-caregiver language in inter
Externí odkaz:
http://arxiv.org/abs/2412.09318
We investigate density fluctuations and scalar-induced gravitational waves (GWs) arising from the production of long-lived solitons and oscillons, which can dominate the early Universe and drive reheating prior to the standard radiation-dominated era
Externí odkaz:
http://arxiv.org/abs/2412.08057
Autor:
Lu, Yen-Ju, Liu, Jing, Thebaud, Thomas, Moro-Velazquez, Laureano, Rastrow, Ariya, Dehak, Najim, Villalba, Jesus
We introduce Condition-Aware Self-Supervised Learning Representation (CA-SSLR), a generalist conditioning model broadly applicable to various speech-processing tasks. Compared to standard fine-tuning methods that optimize for downstream models, CA-SS
Externí odkaz:
http://arxiv.org/abs/2412.04425