mlsanityData to 7 Oct 2026

What makes a good conversation? How controllable attributes affect human judgments

arXiv 1902.08654 ↗First posted 22 Feb 2019cs.CLcited by 140

Not on the map. Why?

Abstract (arXiv)

A good conversation requires balance -- between simplicity and detail; staying on topic and changing it; asking questions and answering them. Although dialogue agents are commonly evaluated via human judgments of overall quality, the relationship between quality and these individual factors is less well-studied. In this work, we examine two controllable neural text generation methods, conditional training and weighted decoding, in order to control four important attributes for chitchat dialogue: repetition, specificity, response-relatedness and question-asking. We conduct a large-scale human evaluation to measure the effect of these control parameters on multi-turn interactive conversations on the PersonaChat task. We provide a detailed analysis of their relationship to high-level aspects of conversation, and show that by controlling combinations of these variables our models obtain clear improvements in human quality judgments.

Key facts. Cited by 140 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Feb 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2020: 1; 2021: 38; 2022: 29; 2023: 30; 2024: 17; 2025: 15; 2026: 10.

140 papers.

Sort:

By area:

  1. 2022LaMDA: Language Models for Dialog Applicationscited by 1,218
  2. 2021DExperts: Decoding-Time Controlled Text Generation with Experts and Anti-Expertscited by 434
  3. 2021FUDGE: Controlled Text Generation With Future Discriminatorscited by 374
  4. 2022SODA: Million-scale Dialogue Distillation with Social Commonsense Contextualizationcited by 166
  5. 2021Recent Advances in Deep Learning Based Dialogue Systems: A Systematic Surveycited by 130
  6. 2022RL with KL penalties is better viewed as Bayesian inferencecited by 130
  7. 2022Productivity Assessment of Neural Code Completioncited by 97
  8. 2020Human-centric Dialog Training via Offline Reinforcement Learningcited by 92
  9. 2021A Comprehensive Assessment of Dialog Evaluation Metricscited by 90
  10. 2023Topical-Chat: Towards Knowledge-Grounded Open-Domain Conversationscited by 88
  11. 2022MISC: A MIxed Strategy-Aware Model Integrating COMET for Emotional Support Conversationcited by 84
  12. 2023Reward-Augmented Decoding: Efficient Controlled Text Generation With a Unidirectional Reward Modelcited by 79
38 more
  1. 2021Anticipating Safety Issues in E2E Conversational AI: Framework and Toolingcited by 77
  2. 2021Towards Facilitating Empathic Conversations in Online Mental Health Support: A Reinforcement Learning Approachcited by 75
  3. 2021Evaluating Attribution in Dialogue Systems: The BEGIN Benchmarkcited by 75
  4. 2021Increasing Faithfulness in Knowledge-Grounded Dialogue with Controllable Featurescited by 74
  5. 2021Building and Evaluating Open-Domain Dialogue Corpora with Clarifying Questionscited by 63
  6. 2022Human Evaluation of Conversations is an Open Problem: comparing the sensitivity of various methods for evaluating dialogue agentscited by 62
  7. 2021Compression, Transduction, and Creation: A Unified Framework for Evaluating Natural Language Generationcited by 60
  8. 2021DynaEval: Unifying Turn and Dialogue Level Evaluationcited by 59
  9. 2021Just Say No: Analyzing the Stance of Neural Dialogue Generation in Offensive Contextscited by 56
  10. 2022LENS: A Learnable Evaluation Metric for Text Simplificationcited by 54
  11. 2021Perturbation CheckLists for Evaluating NLG Evaluation Metricscited by 49
  12. 2023Leveraging Large Language Models to Power Chatbots for Collecting User Self-Reported Datacited by 48
  13. 2023A Comprehensive Analysis of the Effectiveness of Large Language Models as Automatic Dialogue Evaluatorscited by 45
  14. 2021Generate, Prune, Select: A Pipeline for Counterspeech Generation against Online Hate Speechcited by 42
  15. 2021Ask what's missing and what's useful: Improving Clarification Question Generation using Global Knowledgecited by 40
  16. 2022InstructDial: Improving Zero and Few-shot Generalization in Dialogue through Instruction Tuningcited by 39
  17. 2021Multimodal Dialogue Response Generationcited by 32
  18. 2021Controlling Conditional Language Models without Catastrophic Forgettingcited by 32
  19. 2023Elastic Weight Removal for Faithful and Abstractive Dialogue Generationcited by 29
  20. 2021Automatic Evaluation and Moderation of Open-domain Dialogue Systemscited by 28
  21. 2022Semantic Diversity in Dialogue with Natural Language Inferencecited by 28
  22. 2021DeliData: A dataset for deliberation in multi-party problem solvingcited by 26
  23. 2025Enhancing Personalized Multi-Turn Dialogue with Curiosity Rewardcited by 25
  24. 2022Evaluating Mixed-initiative Conversational Search Systems via User Simulationcited by 24
  25. 2022Why is constrained neural language generation particularly challenging?cited by 19
  26. 2022A Simple Contrastive Learning Objective for Alleviating Neural Text Degenerationcited by 18
  27. 2024From One to Many: Expanding the Scope of Toxicity Mitigation in Language Modelscited by 18
  28. 2024IQA-EVAL: Automatic Evaluation of Human-Model Interactive Question Answeringcited by 18
  29. 2023Leftover Lunch: Advantage-based Offline Reinforcement Learning for Language Modelscited by 17
  30. 2024Abstractive Text Summarization: State of the Art, Challenges, and Improvementscited by 17
  31. 2025CtrlDiff: Boosting Large Diffusion Language Models with Dynamic Block Prediction and Controllable Generationcited by 17
  32. 2021Towards Automatic Evaluation of Dialog Systems: A Model-Free Off-Policy Evaluation Approachcited by 16
  33. 2022Measuring and Improving Semantic Diversity of Dialogue Generationcited by 16
  34. 2023NLG Evaluation Metrics Beyond Correlation Analysis: An Empirical Metric Preference Checklistcited by 16
  35. 2021Identifying Untrustworthy Samples: Data Filtering for Open-domain Dialogues with Bayesian Optimizationcited by 15
  36. 2024Assessing and Verifying Task Utility in LLM-Powered Applicationscited by 15
  37. 2022FineD-Eval: Fine-grained Automatic Dialogue-Level Evaluationcited by 13
  38. 2023Towards Designing a ChatGPT Conversational Companion for Elderly Peoplecited by 13

50 most cited of 140 All 140 (JSON)