Cluster-based beam search for pointer-generator chatbot grounded by knowledge. (November 2020)

• Record Type: Journal Article
• Authors: Tam, Yik-Cheung
• Is Part Of: Computer speech & language. Volume 64(2020)
• Publication Date: 2020-11
• Subjects: Pointer generator -- Knowledge-Grounded -- Cluster-Based beam search -- Safe response filtering -- DSTC7; Speech processing systems -- Periodicals; Automatic speech recognition -- Periodicals; Computers -- Periodicals; Linguistics -- Periodicals; Speech-Language Pathology -- Periodicals; Traitement automatique de la parole -- Périodiques; Reconnaissance automatique de la parole -- Périodiques; Automatic speech recognition; Speech processing systems; Electronic journals; Periodicals; 006.454
• DOI: 10.1016/j.csl.2020.101094
• Languages: English
• ISSNs: 0885-2308

Abstract

Highlights: Proposed an end-to-end model to incorporate conversation history and facts with copy mechanism. Proposed a new beam search strategy to improve response diversity driven by K-means clustering. Proposed a language modeling approach to filter out safe responses. Evaluated on DSTC7 and achieved the first place. Post-evaluated system with word-piece tokenization. Abstract: We present an end-to-end approach for knowledge-grounded response generation in Dialog System Technology Challenges 7 (DSTC7). Our system is trained by a pointer generator model, so that an output token in a response can either be generated or copied from conversation history or facts according to a trainable action probability distribution. Furthermore, to minimize generating meaningless responses, we propose using K-means to dynamically cluster and prune semantically similar partial hypotheses at each decoding step under a fixed beam budget. Moreover, we employ a language model to filter meaningless responses. Official evaluation results show that our proposed system achieved the first place in all primary automatic evaluation metrics and the overall human evaluation score.