Why are Sequence-to-Sequence Models So Dull? Understanding the Low-Diversity Problem of Chatbots

Abstract

Diversity is a long-studied topic in information retrieval that usually refers to the requirement that retrieved results should be non-repetitive and cover different aspects. In a conversational setting, an additional dimension of diversity matters: an engaging response generation system should be able to output responses that are diverse and interesting. Sequence-to-sequence (Seq2Seq) models have been shown to be very effective for response generation. However, dialogue responses generated by Seq2Seq models tend to have low diversity. In this paper, we review known sources and existing approaches to this low-diversity problem. We also identify a source of low diversity that has been little studied so far, namely model over-confidence. We sketch several directions for tackling model over-confidence and, hence, the low-diversity problem, including confidence penalties and label smoothing.

References in corpus (13)

• Distilling the Knowledge in a Neural Network
• Sequence to Sequence Learning with Neural Networks
• Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation
• A Neural Conversational Model
• Abstractive Text Summarization Using Sequence-to-Sequence RNNs and Beyond
• A Simple, Fast Diverse Decoding Algorithm for Neural Generation
• Adversarial Learning for Neural Dialogue Generation
• Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders
• A Neural Network Approach to Context-Sensitive Generation of Conversational Responses
• Towards better decoding and language model integration in sequence to sequence models
• Neural Generative Question Answering
• Generating High-Quality and Informative Conversation Responses with Sequence-to-Sequence Models
• Avoiding Echo-Responses in a Retrieval-Based Conversation System