Zobrazeno 1 - 10
of 34
pro vyhledávání: '"He, Wanwei"'
Long-context modeling capabilities are important for large language models (LLMs) in various applications. However, directly training LLMs with long context windows is insufficient to enhance this capability since some training samples do not exhibit
Externí odkaz:
http://arxiv.org/abs/2405.17915
Autor:
Luo, Run, Li, Yunshui, Chen, Longze, He, Wanwei, Lin, Ting-En, Liu, Ziqiang, Zhang, Lei, Song, Zikai, Xia, Xiaobo, Liu, Tongliang, Yang, Min, Hui, Binyuan
The development of large language models (LLMs) has significantly advanced the emergence of large multimodal models (LMMs). While LMMs have achieved tremendous success by promoting the synergy between multimodal comprehension and creation, they often
Externí odkaz:
http://arxiv.org/abs/2405.15232
Autor:
Li, Shujie, Li, Liang, Geng, Ruiying, Yang, Min, Li, Binhua, Yuan, Guanghu, He, Wanwei, Yuan, Shao, Ma, Can, Huang, Fei, Li, Yongbin
Data-to-text (D2T) generation aims to transform structured data into natural language text. Data-to-text pre-training has proved to be powerful in enhancing D2T generation and yields impressive performances. However, previous pre-training methods eit
Externí odkaz:
http://arxiv.org/abs/2401.01183
Autor:
Li, Yunshui, Hui, Binyuan, Yin, Zhaochao, He, Wanwei, Luo, Run, Long, Yuxing, Yang, Min, Huang, Fei, Li, Yongbin
Visually-grounded dialog systems, which integrate multiple modes of communication such as text and visual inputs, have become an increasingly popular area of investigation. However, the absence of a standardized evaluation framework poses a challenge
Externí odkaz:
http://arxiv.org/abs/2309.07387
Autor:
Dai, Yinpei, He, Wanwei, Li, Bowen, Wu, Yuchuan, Cao, Zheng, An, Zhongqi, Sun, Jian, Li, Yongbin
Practical dialog systems need to deal with various knowledge sources, noisy user expressions, and the shortage of annotated data. To better solve the above problems, we propose CGoDial, new challenging and comprehensive Chinese benchmark for multi-do
Externí odkaz:
http://arxiv.org/abs/2211.11617
Recently, pre-training methods have shown remarkable success in task-oriented dialog (TOD) systems. However, most existing pre-trained models for TOD focus on either dialog understanding or dialog generation, but not both. In this paper, we propose S
Externí odkaz:
http://arxiv.org/abs/2209.06664
Autor:
He, Wanwei, Dai, Yinpei, Hui, Binyuan, Yang, Min, Cao, Zheng, Dong, Jianbo, Huang, Fei, Si, Luo, Li, Yongbin
Pre-training methods with contrastive learning objectives have shown remarkable success in dialog understanding tasks. However, current contrastive learning solely considers the self-augmented dialog samples as positive samples and treats all other d
Externí odkaz:
http://arxiv.org/abs/2209.06638
Autor:
He, Wanwei, Dai, Yinpei, Zheng, Yinhe, Wu, Yuchuan, Cao, Zheng, Liu, Dermot, Jiang, Peng, Yang, Min, Huang, Fei, Si, Luo, Sun, Jian, Li, Yongbin
Pre-trained models have proved to be powerful in enhancing task-oriented dialog systems. However, current pre-training methods mainly focus on enhancing dialog understanding and generation tasks while neglecting the exploitation of dialog policy. In
Externí odkaz:
http://arxiv.org/abs/2111.14592
Most existing neural network based task-oriented dialogue systems follow encoder-decoder paradigm, where the decoder purely depends on the source texts to generate a sequence of words, usually suffering from instability and poor readability. Inspired
Externí odkaz:
http://arxiv.org/abs/2106.05830
Autor:
He, Wanwei, Kamely, Mohammad, Wakaruk, Jeremy, Goes, Emanuele C., Korver, Douglas R., Barreda, Daniel R.
Publikováno v:
In Developmental and Comparative Immunology March 2023 140
