Construction Repetition Reduces Information Rate in Dialogue
University of Amsterdam(阿姆斯特丹大学);University of Aberdeen(阿伯丁大学)
Mario Giulianelli, Arabella Sinclair, Raquel Fernández
备注:
In Proceedings of the 2nd Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the 12th International Joint Conference on Natural Language Processing (AACL-IJCNLP 2022)
Speakers repeat constructions frequently in dialogue. Due to their peculiar information-theoretic properties, repetitions can be thought of as a strategy for cost-effective communication. In this study, we focus on the repetition of lexicalised constructions -- i.e., recurring multi-word units -- in English open-domain spoken dialogues. We hypothesise that speakers use construction repetition to mitigate information rate, leading to an overall decrease in utterance information content over the course of a dialogue. We conduct a quantitative analysis, measuring the information content of constructions and that of their containing utterances, estimating information content with an adaptive neural language model. We observe that construction usage lowers the information content of utterances. This facilitating effect (i) increases throughout dialogues, (ii) is boosted by repetition, (iii) grows as a function of repetition frequency and density, and (iv) is stronger for repetitions of referential constructions.
Scaling Conversational Hungarian ASR: The BEA-Dialogue+ Corpus
Department of Telecommunications and Artificial Intelligence(电信与人工智能系);Budapest University of Technology and Economics(布达佩斯技术与经济大学);Speechtex Ltd.(Speechtex公司);ELTE Research Centre for Linguistics(ELTE语言学研究中心)
Copy-Enhanced Heterogeneous Information Learning for Dialogue State Tracking
Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);University of Chinese Academy of Sciences(中国科学院大学);Beijing Unisound Information Technology Co., Ltd(北京云知声信息技术有限公司)
对话状态跟踪(DST)是面向任务的对话系统中的一个重要组成部分,它估计用户在每个对话回合的目标。然而,大多数先前的方法通常遭受以下问题。许多判别模型,特别是端到端(E2 E)模型,很难提取出候选本体中不存在的未知值;而以往的生成模型,虽然可以从话语中提取未知值,但由于忽略了预定义本体的语义信息,导致性能下降。此外,以前的生成模型通常需要手工制作的列表来规范化生成的值。如何整合预定义本体和对话文本(异构文本)的语义信息以生成未知值并提高性能成为一个严峻的挑战。本文提出了一种基于多编码器-解码器的复制增强异构信息学习模型CEDST(Copy Enhanced Heterogeneous Information Learning Model with Multiple Encoder-Decoder for DST),该模型通过从异构文本中复制值,有效地生成包括未知值在内的所有可能值。同时,CEDST通过多个编码器有效地将大的状态空间分解为若干个小的状态空间,并采用多个解码器充分利用这些小的状态空间来产生值。多编解码器架构可以显著提高性能。实验表明,CEDST可以达到最先进的结果在两个数据集和我们构造的数据集与许多未知值。