arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1809.05972cs.CLcs.AI
收藏

通过对抗信息最大化生成信息丰富的多样化会话响应

Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization

Microsoft Research(微软研究院)

Yizhe Zhang, Michel Galley, Jianfeng Gao, Zhe Gan, Xiujun Li, Chris Brockett, Bill Dolan

摘要:

神经会话模型产生的反应往往缺乏信息量和多样性。我们提出了对抗信息最大化(AIM),这是一种对抗性学习策略,可以解决这两个相关但不同的问题。为了促进反应的多样性,我们利用对抗性训练,允许合成反应和真实反应的分布匹配。为了提高信息量,我们的框架显式优化了查询和响应之间的成对互信息的变分下界。自动和人工评估的实证结果表明,我们的方法显着提高信息量和多样性。

英文摘要:

Responses generated by neural conversational models tend to lack informativeness and diversity. We present Adversarial Information Maximization (AIM), an adversarial learning strategy that addresses these two related but distinct problems. To foster response diversity, we leverage adversarial training that allows distributional matching of synthetic and real responses. To improve informativeness, our framework explicitly optimizes a variational lower bound on pairwise mutual information between query and response. Empirical results from automatic and human evaluations demonstrate that our methods significantly boost informativeness and diversity.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑