no code implementations • Findings (EMNLP) 2021 • Qingqing Zhu, Xiuying Chen, Pengfei Wu, Junfei Liu, Dongyan Zhao
Hence, in this paper, we introduce a combination of curriculum learning and knowledge distillation for efficient dialogue generation models, where curriculum learning can help knowledge distillation from data and model aspects.
no code implementations • 18 Apr 2024 • Pengfei Wu, Jiahao Liu, Zhuocheng Gong, Qifan Wang, Jinpeng Li, Jingang Wang, Xunliang Cai, Dongyan Zhao
In this paper, we propose a novel parallel decoding approach, namely \textit{hidden transfer}, which decodes multiple successive tokens simultaneously in a single forward pass.
1 code implementation • 25 Feb 2024 • Yuxuan Wang, Yueqian Wang, Pengfei Wu, Jianxin Liang, Dongyan Zhao, Zilong Zheng
Despite progress in video-language modeling, the computational challenge of interpreting long-form videos in response to task-specific linguistic queries persists, largely due to the complexity of high-dimensional video data and the misalignment between language and visual cues over space and time.
no code implementations • 12 Oct 2023 • Wei Huang, Jixuan Zhou, Fan Gao, Jiajun Lu, Sijia Li, Pengfei Wu, Junting Wang, Hao Zhang, Tianhe Xu
The proposal of SSP inversion method greatly improves the convenience and real--time performance, but the accuracy is not as good as the direct measurement method.
no code implementations • 7 Aug 2023 • Pengfei Wu, Chen Chen, Dexiang Lai, Jian Zhong
Instead of integrating the constraint violation penalty with the reward function, its actor gradients are estimated by a Lagrange advantage function which is derived from two critic systems based on economic reward and violation cost.
no code implementations • 28 Jun 2023 • Ziheng Cheng, Xinmeng Huang, Pengfei Wu, Kun Yuan
When all clients participate in the training process, we demonstrate that incorporating momentum allows FedAvg to converge without relying on the assumption of bounded data heterogeneity even using a constant local learning rate.
1 code implementation • 8 Oct 2021 • Pengfei Wu, Junjie Pan, Chenchang Xu, Junhui Zhang, Lin Wu, Xiang Yin, Zejun Ma
In expressive speech synthesis, there are high requirements for emotion interpretation.
1 code implementation • ACL 2022 • Zhoujun Cheng, Haoyu Dong, Ran Jia, Pengfei Wu, Shi Han, Fan Cheng, Dongmei Zhang
In this paper, we find that the spreadsheet formula, which performs calculations on numerical values in tables, is naturally a strong supervision of numerical reasoning.
no code implementations • 16 Aug 2019 • Peng Chen, Tong Jia, Pengfei Wu, Jianjun Wu, Dongyue Chen
Most existing person re-identification (ReID) methods have good feature representations to distinguish pedestrians with deep convolutional neural network (CNN) and metric learning methods.