Fine-grained Conversational Decoding via Isotropic and Proximal Search

Abstract

General-purpose text decoding approaches are usually adopted for dialogue response generation. Although the quality of the generated responses can be improved with dialogue-specific encoding methods, conversational decoding methods are still under-explored. Inspired by \citet{wu2023learning} that a good dialogue feature space should follow the rules of locality and isotropy, we present a fine-grained conversational decoding method, termed \textit{isotropic and proximal search (IPS)}. Our method is designed to generate the semantic-concentrated response, while still maintaining informativeness and discrimination against the context. Experiments show that our approach outperforms existing decoding strategies in the dialogue field across both automatic and human evaluation metrics. More in-depth analyses further confirm the effectiveness of our approach.

arXiv commentAccepted to EMNLP 2023 Main Conference

References in corpus (8)

• Adam: A Method for Stochastic Optimization
• BERTScore: Evaluating Text Generation with BERT
• DailyDialog: A Manually Labelled Multi-turn Dialogue Dataset
• A Survey on Dialogue Systems: Recent Advances and New Frontiers
• A Contrastive Framework for Neural Text Generation
• Contrastive Search Is What You Need For Neural Text Generation
• Mirostat: A Neural Text Decoding Algorithm that Directly Controls Perplexity
• Learning Locality and Isotropy in Dialogue Modeling