Zobrazeno 1 - 10
of 97
pro vyhledávání: '"Chen, Jiusheng"'
Publikováno v:
Aircraft Engineering and Aerospace Technology, 2023, Vol. 95, Issue 10, pp. 1518-1530.
Externí odkaz:
http://www.emeraldinsight.com/doi/10.1108/AEAT-11-2022-0330
Autor:
Yan, Yu, Hu, Fei, Chen, Jiusheng, Bhendawade, Nikhil, Ye, Ting, Gong, Yeyun, Duan, Nan, Cui, Desheng, Chi, Bingyu, Zhang, Ruofei
Transformer-based models have made tremendous impacts in natural language generation. However the inference speed is a bottleneck due to large model size and intensive computing involved in auto-regressive decoding process. We develop FastSeq framewo
Externí odkaz:
http://arxiv.org/abs/2106.04718
Autor:
Yan, Yu, Chen, Jiusheng, Qi, Weizhen, Bhendawade, Nikhil, Gong, Yeyun, Duan, Nan, Zhang, Ruofei
Transformer model with multi-head attention requires caching intermediate results for efficient inference in generation tasks. However, cache brings new memory-related costs and prevents leveraging larger batch size for faster speed. We propose memor
Externí odkaz:
http://arxiv.org/abs/2105.04779
Autor:
Qi, Weizhen, Gong, Yeyun, Yan, Yu, Xu, Can, Yao, Bolun, Zhou, Bartuer, Cheng, Biao, Jiang, Daxin, Chen, Jiusheng, Zhang, Ruofei, Li, Houqiang, Duan, Nan
Now, the pre-training technique is ubiquitous in natural language processing field. ProphetNet is a pre-training based natural language generation method which shows powerful performance on English text summarization and question generation tasks. In
Externí odkaz:
http://arxiv.org/abs/2104.08006
Autor:
Qi, Weizhen, Gong, Yeyun, Jiao, Jian, Yan, Yu, Chen, Weizhu, Liu, Dayiheng, Tang, Kewen, Li, Houqiang, Chen, Jiusheng, Zhang, Ruofei, Zhou, Ming, Duan, Nan
In this paper, we propose BANG, a new pretraining model to Bridge the gap between Autoregressive (AR) and Non-autoregressive (NAR) Generation. AR and NAR generation can be uniformly regarded as to what extent previous tokens can be attended, and BANG
Externí odkaz:
http://arxiv.org/abs/2012.15525
Publikováno v:
In Measurement 28 February 2024 226
Autor:
Liu, Dayiheng, Yan, Yu, Gong, Yeyun, Qi, Weizhen, Zhang, Hang, Jiao, Jian, Chen, Weizhu, Fu, Jie, Shou, Linjun, Gong, Ming, Wang, Pengcheng, Chen, Jiusheng, Jiang, Daxin, Lv, Jiancheng, Zhang, Ruofei, Wu, Winnie, Zhou, Ming, Duan, Nan
Multi-task benchmarks such as GLUE and SuperGLUE have driven great progress of pretraining and transfer learning in Natural Language Processing (NLP). These benchmarks mostly focus on a range of Natural Language Understanding (NLU) tasks, without con
Externí odkaz:
http://arxiv.org/abs/2011.11928
Tell Me How to Ask Again: Question Data Augmentation with Controllable Rewriting in Continuous Space
Autor:
Liu, Dayiheng, Gong, Yeyun, Fu, Jie, Yan, Yu, Chen, Jiusheng, Lv, Jiancheng, Duan, Nan, Zhou, Ming
In this paper, we propose a novel data augmentation method, referred to as Controllable Rewriting based Question Data Augmentation (CRQDA), for machine reading comprehension (MRC), question generation, and question-answering natural language inferenc
Externí odkaz:
http://arxiv.org/abs/2010.01475
Autor:
Liu, Dayiheng, Gong, Yeyun, Fu, Jie, Yan, Yu, Chen, Jiusheng, Jiang, Daxin, Lv, Jiancheng, Duan, Nan
Reading long documents to answer open-domain questions remains challenging in natural language understanding. In this paper, we introduce a new model, called RikiNet, which reads Wikipedia pages for natural question answering. RikiNet contains a dyna
Externí odkaz:
http://arxiv.org/abs/2004.14560