Open-Domain Conversational Agents: Current Progress, Open Problems, and Future Directions
Not on the map. Why?
Abstract (arXiv)
We present our view of what is necessary to build an engaging open-domain conversational agent: covering the qualities of such an agent, the pieces of the puzzle that have been built so far, and the gaping holes we have not filled yet. We present a biased view, focusing on work done by our own group, while citing related work in each area. In particular, we discuss in detail the properties of continual learning, providing engaging content, and being well-behaved -- and how to measure success in providing them. We end with a discussion of our experience and learnings, and our recommendations to the community.
Key facts. Cited by 34 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Jun 2020.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 8; 2022: 9; 2023: 8; 2024: 1; 2025: 8.
34 papers.
Sort:
By area:
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
- 2022Survey of Hallucination in Natural Language Generationcited by 2,056
- 2022BlenderBot 3: a deployed conversational agent that continually learns to responsibly engagecited by 197
- 2022Human Evaluation of Conversations is an Open Problem: comparing the sensitivity of various methods for evaluating dialogue agentscited by 62
- 2023Multi-Party Chat: Conversational Agents in Group Settings with Humans and Modelscited by 53
- 2023AART: AI-Assisted Red-Teaming with Diverse Data Generation for New LLM-powered Applicationscited by 53
- 2022Neural Approaches to Conversational Information Retrievalcited by 41
- 2021SaFeRDialogues: Taking Feedback Gracefully after Conversational Safety Failurescited by 37
- 2021Rome was built in 1776: A Case Study on Factual Correctness in Knowledge-Grounded Response Generationcited by 29
- 2021Investigating Robustness of Dialog Models to Popular Figurative Language Constructscited by 19
- 2025Compute as Teacher: Turning Inference Compute Into Reference-Free Supervisioncited by 17
- 2022Multimodal Conversational AI: A Survey of Datasets and Approachescited by 16
22 more
- 2023History-Aware Hierarchical Transformer for Multi-session Open-domain Dialogue Systemcited by 15
- 2022BotsTalk: Machine-sourced Framework for Automatic Curation of Large-scale Multi-skill Dialogue Datasetscited by 13
- 2022Learning from data in the mixed adversarial non-adversarial case: Finding the helpers and ignoring the trollscited by 11
- 2023Improving Open Language Models by Learning from Organic Interactionscited by 10
- 2021Studying Catastrophic Forgetting in Neural Ranking Modelscited by 9
- 2025Min-p, Max Exaggeration: A Critical Analysis of Min-p Sampling in Language Modelscited by 9
- 2021Bilateral Personalized Dialogue Generation with Contrastive Learningcited by 8
- 2021Deep Transfer Learning & Beyond: Transformer Language Models in Information Systems Researchcited by 7
- 2025Position: Machine Learning Conferences Should Establish a "Refutations and Critiques" Trackcited by 7
- 2023When Crowd Meets Persona: Creating a Large-Scale Open-Domain Persona Dialogue Corpuscited by 5
- 2022Neural Generation Meets Real People: Building a Social, Informative Open-Domain Dialogue Agentcited by 4
- 2023The Open-domain Paradox for Chatbots: Common Ground as the Basis for Human-like Dialoguecited by 4
- 2024Chitchat as Interference: Adding User Backstories to Task-Oriented Dialoguescited by 4
- 2021Teach Me What to Say and I Will Learn What to Pick: Unsupervised Knowledge Selection Through Response Generation with Pretrained Generative Modelscited by 3
- 2022Controllable Factuality in Document-Grounded Dialog Systems Using a Noisy Channel Modelcited by 3
- 2025Correlating and Predicting Human Evaluations of Language Models from Natural Language Processing Benchmarkscited by 3
- 2021An Approach to Inference-Driven Dialogue Management within a Social Chatbotcited by 2
- 2025Prior Lessons of Incremental Dialogue and Robot Action Management for the Age of Language Modelscited by 2
- 2023Enhancing Task-Oriented Dialogues with Chitchat: a Comparative Study Based on Lexical Diversity and Divergencecited by 1
- 2025ISCA: A Framework for Interview-Style Conversational Agentscited by 0
- 2025Towards Automated Error Discovery: A Study in Conversational AIcited by 0
- 2025Telling Speculative Stories to Help Humans Imagine the Harms of Healthcare AIcited by 0
- 2025Telling Speculative Stories to Help Humans Imagine the Harms of Healthcare AI16 Oct 2025
- 2025Compute as Teacher: Turning Inference Compute Into Reference-Free Supervision17 Sep 2025
- 2025Towards Automated Error Discovery: A Study in Conversational AI13 Sep 2025
- 2025ISCA: A Framework for Interview-Style Conversational Agents20 Aug 2025
- 2025Position: Machine Learning Conferences Should Establish a "Refutations and Critiques" Track24 Jun 2025
- 2025Min-p, Max Exaggeration: A Critical Analysis of Min-p Sampling in Language Models16 Jun 2025
- 2025Correlating and Predicting Human Evaluations of Language Models from Natural Language Processing Benchmarks24 Feb 2025
- 2025Prior Lessons of Incremental Dialogue and Robot Action Management for the Age of Language Models1 Jan 2025
- 2024Chitchat as Interference: Adding User Backstories to Task-Oriented Dialogues23 Feb 2024
- 2023Enhancing Task-Oriented Dialogues with Chitchat: a Comparative Study Based on Lexical Diversity and Divergence23 Nov 2023
- 2023AART: AI-Assisted Red-Teaming with Diverse Data Generation for New LLM-powered Applications14 Nov 2023
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Models18 Jul 2023
22 more
- 2023Improving Open Language Models by Learning from Organic Interactions7 Jun 2023
- 2023Multi-Party Chat: Conversational Agents in Group Settings with Humans and Models26 Apr 2023
- 2023When Crowd Meets Persona: Creating a Large-Scale Open-Domain Persona Dialogue Corpus1 Apr 2023
- 2023The Open-domain Paradox for Chatbots: Common Ground as the Basis for Human-like Dialogue21 Mar 2023
- 2023History-Aware Hierarchical Transformer for Multi-session Open-domain Dialogue System2 Feb 2023
- 2022Controllable Factuality in Document-Grounded Dialog Systems Using a Noisy Channel Model31 Oct 2022
- 2022BotsTalk: Machine-sourced Framework for Automatic Curation of Large-scale Multi-skill Dialogue Datasets23 Oct 2022
- 2022Learning from data in the mixed adversarial non-adversarial case: Finding the helpers and ignoring the trolls5 Aug 2022
- 2022BlenderBot 3: a deployed conversational agent that continually learns to responsibly engage5 Aug 2022
- 2022Neural Generation Meets Real People: Building a Social, Informative Open-Domain Dialogue Agent25 Jul 2022
- 2022Multimodal Conversational AI: A Survey of Datasets and Approaches13 May 2022
- 2022Survey of Hallucination in Natural Language Generation8 Feb 2022
- 2022Neural Approaches to Conversational Information Retrieval13 Jan 2022
- 2022Human Evaluation of Conversations is an Open Problem: comparing the sensitivity of various methods for evaluating dialogue agents12 Jan 2022
- 2021An Approach to Inference-Driven Dialogue Management within a Social Chatbot31 Oct 2021
- 2021Deep Transfer Learning & Beyond: Transformer Language Models in Information Systems Research18 Oct 2021
- 2021SaFeRDialogues: Taking Feedback Gracefully after Conversational Safety Failures14 Oct 2021
- 2021Rome was built in 1776: A Case Study on Factual Correctness in Knowledge-Grounded Response Generation11 Oct 2021
- 2021Teach Me What to Say and I Will Learn What to Pick: Unsupervised Knowledge Selection Through Response Generation with Pretrained Generative Models5 Oct 2021
- 2021Investigating Robustness of Dialog Models to Popular Figurative Language Constructs1 Oct 2021
- 2021Bilateral Personalized Dialogue Generation with Contrastive Learning15 Jun 2021
- 2021Studying Catastrophic Forgetting in Neural Ranking Models18 Jan 2021
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
- 2022Survey of Hallucination in Natural Language Generationcited by 2,056
- 2022BlenderBot 3: a deployed conversational agent that continually learns to responsibly engagecited by 197
- 2022Human Evaluation of Conversations is an Open Problem: comparing the sensitivity of various methods for evaluating dialogue agentscited by 62
- 2022Neural Approaches to Conversational Information Retrievalcited by 41
- 2021SaFeRDialogues: Taking Feedback Gracefully after Conversational Safety Failurescited by 37
- 2021Rome was built in 1776: A Case Study on Factual Correctness in Knowledge-Grounded Response Generationcited by 29
- 2021Investigating Robustness of Dialog Models to Popular Figurative Language Constructscited by 19
- 2025Compute as Teacher: Turning Inference Compute Into Reference-Free Supervisioncited by 17
- 2022Multimodal Conversational AI: A Survey of Datasets and Approachescited by 16
- 2023History-Aware Hierarchical Transformer for Multi-session Open-domain Dialogue Systemcited by 15
- 2022BotsTalk: Machine-sourced Framework for Automatic Curation of Large-scale Multi-skill Dialogue Datasetscited by 13
12 most cited of 32 All 32 (JSON)
- 2025Telling Speculative Stories to Help Humans Imagine the Harms of Healthcare AI16 Oct 2025
- 2025Compute as Teacher: Turning Inference Compute Into Reference-Free Supervision17 Sep 2025
- 2025Towards Automated Error Discovery: A Study in Conversational AI13 Sep 2025
- 2025ISCA: A Framework for Interview-Style Conversational Agents20 Aug 2025
- 2025Position: Machine Learning Conferences Should Establish a "Refutations and Critiques" Track24 Jun 2025
- 2025Min-p, Max Exaggeration: A Critical Analysis of Min-p Sampling in Language Models16 Jun 2025
- 2025Correlating and Predicting Human Evaluations of Language Models from Natural Language Processing Benchmarks24 Feb 2025
- 2025Prior Lessons of Incremental Dialogue and Robot Action Management for the Age of Language Models1 Jan 2025
- 2024Chitchat as Interference: Adding User Backstories to Task-Oriented Dialogues23 Feb 2024
- 2023Enhancing Task-Oriented Dialogues with Chitchat: a Comparative Study Based on Lexical Diversity and Divergence23 Nov 2023
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Models18 Jul 2023
- 2023Improving Open Language Models by Learning from Organic Interactions7 Jun 2023
12 newest of 32 All 32 (JSON)