Neural personalized response generation as domain adaptation

Neural personalized response generation as domain adaptation
复制标题

作为领域适应的神经个性化响应生成

DOI:
10.1007/s11280-018-0598-6
复制
发表时间:
2017-01
期刊:
World Wide Web
影响因子:
--
通讯作者:
Ting Liu
Ting Liu
中科院分区:
其他
文献类型:
--
作者:
Wei-Nan Zhang;Qingfu Zhu;Yifa Wang;Yanyan Zhao;Ting Liu

文献摘要

参考文献

被引文献

相似文献

训练会话机器人的个性化响应生成模型的最关键问题之一是缺乏大规模的个人会话数据。为了解决这个问题,我们提出了一种两阶段的方法,即初始化然后适应,首先在大规模会话数据中预训练优化的RNN编码器-解码器模型(LTS模型)以生成一般响应,然后在小规模个人会话数据中微调模型以生成个性化响应。对于评估,我们提出了一种新的人类辅助方法,它可以被看作是一个准图灵测试,以评估个性化的响应生成模型的性能。实验结果表明,所提出的个性化响应生成模型在自动评估、离线人工判断和准图灵测试方面优于语言模型个性化和基于人物的神经会话生成方法.
One of the most crucial problem on training personalized response generation models for conversational robots is the lack of large scale personal conversation data. To address the problem, we propose a two-phase approach, namelyinitialization then adaptation, to first pre-train an optimized RNN encoder-decoder model (LTSmodel) in a large scale conversational data for general response generation and then fine-tune the model in a small scale personal conversation data to generate personalized responses. For evaluation, we propose a novel human aided method, which can be seen as a quasi-Turing test, to evaluate the performance of the personalized response generation models. Experimental results show that the proposed personalized response generation model outperforms the state-of-the-art approaches to language model personalization and persona-based neural conversation generation on the automatic evaluation, offline human judgment and the quasi-Turing test.
DOI: --
发表时间: 2012-07
期刊: --
影响因子: --
作者:
Rafael E. Banchs;Haizhou Li
通讯作者: Rafael E. Banchs;Haizhou Li
DOI: --
发表时间: 2016-10
期刊: ArXiv
影响因子: --
作者:
Kaixiang Mo;Shuangyin Li;Yu Zhang;Jiajun Li;Qiang Yang
通讯作者: Kaixiang Mo;Shuangyin Li;Yu Zhang;Jiajun Li;Qiang Yang
DOI: 10.3115/1706269.1706277
发表时间: 2006-07
期刊: --
影响因子: --
作者:
Amy Isard;Carsten Brockmann;J. Oberlander
通讯作者: Amy Isard;Carsten Brockmann;J. Oberlander
DOI: --
发表时间: 2015-02
期刊: ArXiv
影响因子: --
作者:
Xiang Zhang-;Yann LeCun
通讯作者: Xiang Zhang-;Yann LeCun
DOI: 10.1145/3077136.3080706
发表时间: 2017-08
期刊: Proceedings of the 40th International ACM SIGIR Conference on Research and Development in Information Retrieval
影响因子: --
作者:
Min Yang;Zhou Zhao;Wei Zhao;Xiaojun Chen;Jia Zhu;Lianqiang Zhou;Zigang Cao
通讯作者: Min Yang;Zhou Zhao;Wei Zhao;Xiaojun Chen;Jia Zhu;Lianqiang Zhou;Zigang Cao