A Dataset for Document Grounded Conversations

By Kangyan Zhou, Shrimai Prabhumoye, Alan W. Black. 2018

249 citations recorded by the source corpus.

This paper introduces a document grounded dataset for conversations. We define “Document Grounded Conversations” as conversations that are about the contents of a specified document. In this dataset the specified documents were Wikipedia articles about popular movies. The dataset contains 4112 conversations with an average of 21.43 turns per conversation. This positions this dataset to not only provide a relevant chat history while generating responses but also provide a source of information that the models could use. We describe two neural architectures that provide benchmark performance on the task of generating the next response. We also evaluate our models for engagement and fluency, and find that the information from the document helps in generating more engaging and fluent responses.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Natural Language Processing, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Computer Vision and Pattern Recognition

In Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

CoQA: A Conversational Question Answering Challenge (2018, background): “Zhou et al. (2018) is another dialogue dataset based on a single movie-related Wikipedia article, in which two workers are asked to chat about the content.”
Hindsight: Posterior-guided training of retrievers for improved open-ended generation (2021, background): “, 2018) or engage in informative conversations (Dinan et al., 2019; Zhou et al., 2018).”
Retrieval Augmentation Reduces Hallucination in Conversation (2021, methodology): “, 2019b) and CMU Document Grounded Conversations (CMU_DoG) (Zhou et al., 2018).”
Survey of Hallucination in Natural Language Generation (2022, background): “In addition to PersonaChat-related research, the knowledge-grounded dialogue (KGD) task in the open-domain requires the model to generate informative responses with the help of an external knowledge graph (KG) or knowledge corpus [32, 238].”
Internet-Augmented Dialogue Generation (2021, background): “In dialogue, knowledge grounding is becoming more popular an area, with several datasets developed to study it (Zhou et al., 2018; Dinan et al., 2019; Ghazvininejad et al., 2018; Gopalakrishnan et al., 2019; Galetzka et al., 2020).”
Reason first, then respond: Modular Generation for Knowledge-infused Dialogue (2021, background): “Knowledgegrounded open-domain dialogue datasets (Dinan et al., 2019; Komeili et al., 2021; Zhou et al., 2018; Gopalakrishnan et al., 2019) foster the research and development of knowledge-aware generative dialogue models.”
On the Origin of Hallucinations in Conversational Models: Is it the Datasets or the Models? (2022, methodology): “…frameworks such as BEGIN (Dziri et al., 2021b) and AIS (Rashkin et al., 2021a), we annotate responses from the three widely-used knowledge-grounded conversational benchmarks: Wizard of Wikipedia (Dinan et al., 2018), CMU-D O G (Zhou et al., 2018) and T OPI - CAL C HAT (Gopalakrishnan et al., 2019).”
A Large-Scale Chinese Short-Text Conversation Dataset (2020, background): “These datasets are built for advanced dialogue tasks, such as wizard of wikipedia (WoW) [6] / document grounded conversations (DoG) [40] for knowledge grounded dialogue generation, PERSONA-CHAT [37] for persona enhanced dialogue generation, and DailyDialog [13] for emotional conversation generation.”
Are Pre-trained Language Models Knowledgeable to Ground Open Domain Dialogues? (2020, background): “The advantage of DialogGPTfinetune over the baselines is even bigger on CMU DoG, indicating that pre-training is more crucial when training data is small in size.”
Case Study: Deontological Ethics in NLP (2020, background): “, 1998), style (Niu and Bansal, 2018), content (Zhou et al., 2018; Zhao et al., 2020; Dinan et al., 2019) and persona (Zhang et al.”
MMDialog: A Large-scale Multi-turn Dialogue Dataset Towards Multi-modal Open-domain Conversation (2022, background): “Empowering machines to converse like humans is a long-cherished goal of AI community, and there is growing interest in developing open-domain conversational agents (Li et al., 2017; Gao et al., 2018; Ghazvininejad et al., 2018; Zhou et al., 2018a).”
Grounded Conversation Generation as Guided Traverses in Commonsense Knowledge Graphs (2019, methodology): “, 2018a), or background documents (Zhou et al., 2018b).”
Topical-Chat: Towards Knowledge-Grounded Open-Domain Conversations (2019, background): “[6] released a dataset of∼4K conversations where Wikipedia articles about 30 movies served as the knowledge base.”
A Bichannel Transformer with Context Encoding for Document-Driven Conversation Generation in Social Media (2020, methodology): “We evaluate the proposed model on the dataset CMU_DoG (CMU document-grounded conversations) for document-driven conversations [24].”
Sequential Latent Knowledge Selection for Knowledge-Grounded Dialogue (2020, methodology): “…conversations have been studied much including collecting new datasets (Qin et al., 2019; Zhang et al., 2018; Ghazvininejad et al., 2018; Zhou et al., 2018; Dinan et al., 2019; Moghe et al., 2018) or developing new models (Lian et al., 2019; Li et al., 2019b; Yavuz et al., 2019; Zhao et…”
CoLV: A Collaborative Latent Variable Model for Knowledge-Grounded Dialogue Generation (2021, background): “Knowledge-grounded dialogue generation (Liu et al., 2018; Zhou et al., 2018a; Lian et al., 2019; Tian et al., 2020), which utilizes external knowledge to enhance conversation backgrounds, has achieved promising performance.”
Read before Generate! Faithful Long Form Question Answering with Machine Reading (2022, background): “Zhou et al. (2018) introduced a document grounded dataset for text conversations, and Wu et al. (2021) proposed to extract lexical control phrases to do controllable grounded response generation, while Zhang et al. (2021) jointly trained a retriever and generator so that annotated text reference…”