Explore models of dialog in this comprehensive lecture from CMU's Neural Networks for NLP course. Delve into chat-based and task-based dialog systems, covering generation-based models, neural approaches for response generation, and techniques to improve global coherence. Examine methods for addressing diversity in evaluation, infusing personality into dialog agents, and implementing retrieval-based chat systems. Learn about traditional task-completion dialog frameworks and how neural networks can be applied to natural language understanding, dialog state tracking, and language generation from dialog states. Gain insights into cutting-edge research and practical applications in the field of conversational AI.
Overview
Syllabus
Intro
Types of Dialog
Two Paradigms
Generation-based Models (Ritter et al. 2011)
Neural Models for Dialog Response Generation
Dialog More Dependent on Global Coherence
One Solution: Use Standard Architecture w/ More Context
Hierarchical Encoder- decoder Model (Serban et al. 2016)
Discourse-level VAE Model (Zhao et al. 2017)
Diversity is a Problem for Evaluation!
Using Multiple References with Human Evaluation Scores (Galley et al. 2015)
Learning to Evaluate
Dialog Agents should have Personality
Personality Infused Dialog (Mairesse et al. 2007)
Dialog Response Retrieval
Retrieval-based Chat (Lee et al. 2009)
Neural Response Retrieval (Nio et al. 2014)
Smart Reply for Email Retrieval (Kannan et al. 2016)
Chat vs. Task Completion
Traditional Task-completion Dialog Framework
NLU (for Slot Filling) w/ Neural Nets (Mesnil et al. 2015)
Dialog State Tracking
Language Generation from Dialog State w/ Neural Nets (Wen et al. 2015)
Taught by
Graham Neubig
