Why Do Neural Dialog Systems Generate Short and Meaningless Replies? A Comparison between Dialog and Translation

2017/12/06 by Bolin Wei, Shuai Lu, Wei, Bolin +11 · 3 citations
Computer Science · #Topic Modeling #Natural Language Processing Techniques #Multimodal Machine Learning Applications

paper · pdf · doi:10.48550/arxiv.1712.02250

Abstract

This paper addresses the question: Why do neural dialog systems generate short and meaningless replies? We conjecture that, in a dialog system, an utterance may have multiple equally plausible replies, causing the deficiency of neural networks in the dialog application. We propose a systematic way to mimic the dialog scenario in a machine translation system, and manage to reproduce the phenomenon of generating short and less meaningful sentences in the translation setting, showing evidence of our conjecture.

Citations

• Diverse Beam Search: Decoding Diverse Solutions from Neural Sequence Models

Cited by

• Learning Dynamic Scene Reconstruction with Sinusoidal Geometric Priors
• SirenPose: Dynamic Scene Reconstruction via Geometric Supervision
• Policy-Driven Neural Response Generation for Knowledge-Grounded Dialogue Systems

Related

• Neural Conversation Models and How to Rein Them in: A Survey of Failures and Fixes
• Dialogue as Autocommunication - On Interactions with Large Language Models
• Do Neural Dialog Systems Use the Conversation History Effectively? An Empirical Study
• Chat as Expected: Learning to Manipulate Black-box Neural Dialogue Models
• End-to-end Conversation Modeling Track in DSTC6