Summary

In this document, we present the tables summaried form the papers cited in the paper list.

Multimodal Dialogue

Paper	Dataset	Visual Encoder	Language Model	Encoder	Decoder
Towards Building Large Scale Multimodal Domain-Aware Conversation Systems	MMD	VGGNet-16	Bidirectional RNN network with GRU	GRU (HRED)	GRU + Attention

Paper	Issue	Dataset	Language Model	Encoder	Decoder
The JDDC Corpus: A Large-Scale Multi-Turn Chinese Dialogue Dataset for E-commerce Customer Service		JDDC	Multimodal Dialogue Dense Retriever; Multimodal Hierarchical Encoder Decoder
Towards Scalable Multi-domain Conversational Agents: The Schema-Guided Dialogue Dataset	Benchmark; Zero-shot DST	SGD	BERT-DST (Schema Embedding Module + State Update Module)	BERT	Softmax
CrossWOZ: A Large-Scale Chinese Cross-Domain Task-Oriented Dialogue Dataset		CrossWOZ		BERTNLU-context
STAR: A Schema-Guided Dialog Dataset for Transfer Learning	Zero-shot generalization across tasks and domains; Schema = task specification ≠ SGD	STAR	BERT (language + schema; classification); BERT (for generation)	BERT+classifier (for generation)	Fine-tune GPT-2 (for generation)
RiSAWOZ: A Large-Scale Multi-Domain Wizard-of-Oz Dataset with Rich Semantic Annotations for Task-Oriented Dialogue Modeling		RiSAWOZ	BERT + context (for NLU); TRADE + MLCSG (for DST); DAMD (for Context2Text Generation); e2e-coref (for Coref Resolution); GECOR (for Unified generative ellipsis + Coref Resolution)
Action-Based Conversations Dataset: A Corpus for Building More In-Depth Task-Oriented Dialogue Systems		ABCD
Adding Chit-Chats to Enhance Task-Oriented Dialogues	Integrate both types of systems (task-oriented systems + open-domain chatbots)	ACCENTOR

Paper	Dataset	Domain	Size

|