Zobrazeno 1 - 10
of 14 118
pro vyhledávání: '"SONG, Yan"'
Deep Learning (DL) libraries, such as PyTorch, are widely used for building and deploying DL models on various hardware platforms. Meanwhile, they are found to contain bugs that lead to incorrect calculation results and cause issues like non-converge
Externí odkaz:
http://arxiv.org/abs/2412.06430
Autor:
Wang, Jun, Fang, Meng, Wan, Ziyu, Wen, Muning, Zhu, Jiachen, Liu, Anjie, Gong, Ziqin, Song, Yan, Chen, Lei, Ni, Lionel M., Yang, Linyi, Wen, Ying, Zhang, Weinan
In this technical report, we introduce OpenR, an open-source framework designed to integrate key components for enhancing the reasoning capabilities of large language models (LLMs). OpenR unifies data acquisition, reinforcement learning training (bot
Externí odkaz:
http://arxiv.org/abs/2410.09671
Reanalysis data, such as ERA5, provide a comprehensive and detailed representation of the Earth's system by assimilating observations into climate models. While crucial for climate research, they pose significant challenges in terms of generation, st
Externí odkaz:
http://arxiv.org/abs/2410.08945
Autor:
Yan, Xue, Song, Yan, Feng, Xidong, Yang, Mengyue, Zhang, Haifeng, Ammar, Haitham Bou, Wang, Jun
In sequential decision-making (SDM) tasks, methods like reinforcement learning (RL) and heuristic search have made notable advances in specific cases. However, they often require extensive exploration and face challenges in generalizing across divers
Externí odkaz:
http://arxiv.org/abs/2410.07927
Autor:
Jiang, Zhaohui, Feng, Xuening, Weng, Paul, Zhu, Yifei, Song, Yan, Zhou, Tianze, Hu, Yujing, Lv, Tangjie, Fan, Changjie
In practice, reinforcement learning (RL) agents are often trained with a possibly imperfect proxy reward function, which may lead to a human-agent alignment issue (i.e., the learned policy either converges to non-optimal performance with low cumulati
Externí odkaz:
http://arxiv.org/abs/2410.05782
A significant challenge in sound event detection (SED) is the effective utilization of unlabeled data, given the limited availability of labeled data due to high annotation costs. Semi-supervised algorithms rely on labeled data to learn from unlabele
Externí odkaz:
http://arxiv.org/abs/2409.17656
We construct entire curves in projective spaces that are simultaneously frequently hypercyclic for translations along countably many specified directions, while preserving optimal slow growth rates. Moreover, we demonstrate that there do not exist en
Externí odkaz:
http://arxiv.org/abs/2409.08048
Autor:
Chen, Yihao, Wu, Haochen, Jiang, Nan, Xia, Xiang, Gu, Qing, Hao, Yunqi, Cai, Pengfei, Guan, Yu, Wang, Jialong, Xie, Weilin, Fang, Lei, Fang, Sian, Song, Yan, Guo, Wu, Liu, Lin, Xu, Minqiang
This paper describes the USTC-KXDIGIT system submitted to the ASVspoof5 Challenge for Track 1 (speech deepfake detection) and Track 2 (spoofing-robust automatic speaker verification, SASV). Track 1 showcases a diverse range of technical qualities fro
Externí odkaz:
http://arxiv.org/abs/2409.01695
Sound event detection (SED) methods that leverage a large pre-trained Transformer encoder network have shown promising performance in recent DCASE challenges. However, they still rely on an RNN-based context network to model temporal dependencies, la
Externí odkaz:
http://arxiv.org/abs/2408.08673