mlsanityData to 7 Oct 2026

ProsocialDialog: A Prosocial Backbone for Conversational Agents

arXiv 2205.12688 ↗First posted 25 May 2022cs.CLcited by 109

On the map: LLM agents and tool use → Social simulation and multi-agent debate

Abstract (arXiv)

Most existing dialogue systems fail to respond properly to potentially unsafe user utterances by either ignoring or passively agreeing with them. To address this issue, we introduce ProsocialDialog, the first large-scale multi-turn dialogue dataset to teach conversational agents to respond to problematic content following social norms. Covering diverse unethical, problematic, biased, and toxic situations, ProsocialDialog contains responses that encourage prosocial behavior, grounded in commonsense social rules (i.e., rules-of-thumb, RoTs). Created via a human-AI collaborative framework, ProsocialDialog consists of 58K dialogues, with 331K utterances, 160K unique RoTs, and 497K dialogue safety labels accompanied by free-form rationales. With this dataset, we introduce a dialogue safety detection module, Canary, capable of generating RoTs given conversational context, and a socially-informed dialogue agent, Prost. Empirical results show that Prost generates more socially acceptable dialogues compared to other state-of-the-art language and dialogue models in both in-domain and out-of-domain settings. Additionally, Canary effectively guides conversational agents and off-the-shelf language models to generate significantly more prosocial responses. Our work highlights the promise and importance of creating and steering conversational AI to be socially responsible.

Key facts. Cited by 109 papers in the map (data through 7 Oct 2026). Placed in Social simulation and multi-agent debate, one of 16 directions in LLM agents and tool use. Builds on 43 of its 87 references that are in the map. First posted on arXiv 25 May 2022.

Builds on

43 of 87 references.

References · 43

  1. 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
  2. 2014Adam: A Method for Stochastic Optimizationcited by 36,960
  3. 2020Language Models are Few-Shot Learnerscited by 32,199
  4. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  5. 2019BART: Denoising Sequence-to-Sequence Pre-training for Natural Language Generation, Translation, and Comprehensioncited by 5,874
  6. 2022OPT: Open Pre-trained Transformer Language Modelscited by 3,464
  7. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,235
  8. 2022LaMDA: Language Models for Dialog Applicationscited by 1,218
  9. 2021Ethical and social risks of harm from Language Modelscited by 957
  10. 2019DialoGPT: Large-Scale Generative Pre-training for Conversational Response Generationcited by 842
  11. 2018Personalizing Dialogue Agents: I have a dog, do you have pets too?cited by 820
  12. 2020Aligning AI With Shared Human Valuescited by 706
31 more
  1. 2017DailyDialog: A Manually Labelled Multi-turn Dialogue Datasetcited by 649
  2. 2020Recipes for building an open-domain chatbotcited by 593
  3. 2018Wizard of Wikipedia: Knowledge-Powered Conversational agentscited by 530
  4. 2020GoEmotions: A Dataset of Fine-Grained Emotionscited by 479
  5. 2016A Corpus and Evaluation Framework for Deeper Understanding of Commonsense Storiescited by 413
  6. 2019Social Bias Frames: Reasoning about Social and Power Implications of Languagecited by 377
  7. 2021Towards Emotional Support Dialog Systemscited by 294
  8. 2021Symbolic Knowledge Distillation: from General Language Models to Commonsense Modelscited by 270
  9. 2021Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detectioncited by 258
  10. 2020Social Chemistry 101: Learning to Reason about Social and Moral Normscited by 253
  11. 2021Internet-Augmented Dialogue Generationcited by 217
  12. 2019Build it Break it Fix it for Dialogue Safety: Robustness from Adversarial Human Attackcited by 176
  13. 2018The Design and Implementation of XiaoIce, an Empathetic Social Chatbotcited by 172
  14. 2020Recipes for Safety in Open-domain Chatbotscited by 168
  15. 2017ParlAI: A Dialog Research Software Platformcited by 139
  16. 2020Can You Put it All Together: Evaluating Conversational Agents' Ability to Blend Skillscited by 130
  17. 2021Challenges in Automated Debiasing for Toxic Language Detectioncited by 95
  18. 2022The Moral Integrity Corpus: A Benchmark for Ethical Dialogue Systemscited by 92
  19. 2021Understanding by Understanding Not: Modeling Negation in Language Modelscited by 73
  20. 2021On the Safety of Conversational Models: Taxonomy, Dataset, and Benchmarkcited by 73
  21. 2021Few-Shot Bot: Prompt-Based Learning for Dialogue Systemscited by 64
  22. 2018Towards Exploiting Background Knowledge for Building Conversation Systemscited by 57
  23. 2021Just Say No: Analyzing the Stance of Neural Dialogue Generation in Offensive Contextscited by 56
  24. 2020A Taxonomy of Empathetic Response Intents in Human Social Conversationscited by 51
  25. 2021Ethical-Advice Taker: Do Language Models Understand Natural Language Interventions?cited by 46
  26. 2021SaFeRDialogues: Taking Feedback Gracefully after Conversational Safety Failurescited by 37
  27. 2020Towards Unified Dialogue System Evaluation: A Comprehensive Analysis of Current Evaluation Protocolscited by 35
  28. 2021ValueNet: A New Dataset for Human Value Driven Dialogue Systemcited by 34
  29. 2021A Word on Machine Ethics: A Response to Jiang et al. (2021)cited by 23
  30. 2022The AI Index 2022 Annual Reportcited by 18
  31. 2021Probing Commonsense Explanation in Dialogue Response Generationcited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 5; 2023: 32; 2024: 36; 2025: 22; 2026: 14.

109 papers.

Sort:

By area:

  1. 2023Bias and Fairness in Large Language Models: A Surveycited by 626
  2. 2024OR-Bench: An Over-Refusal Benchmark for Large Language Modelscited by 222
  3. 2023Can LLMs Keep a Secret? Testing Privacy Implications of Language Models via Contextual Integrity Theorycited by 218
  4. 2023FLASK: Fine-grained Language Model Evaluation based on Alignment Skill Setscited by 198
  5. 2022SODA: Million-scale Dialogue Distillation with Social Commonsense Contextualizationcited by 166
  6. 2025Large Language Models Often Know When They Are Being Evaluatedcited by 105
  7. 2024The Language Barrier: Dissecting Safety Challenges of LLMs in Multilingual Contextscited by 92
  8. 2023Red teaming ChatGPT via Jailbreaking: Bias, Robustness, Reliability and Toxicitycited by 86
  9. 2023A Survey on Proactive Dialogue Systems: Problems, Methods, and Prospectscited by 66
  10. 2024CoSafe: Evaluating Large Language Model Safety in Multi-Turn Dialogue Coreferencecited by 51
  11. 2024Towards Human-centered Proactive Conversational Agentscited by 48
  12. 2023Conversation Chronicles: Towards Diverse Temporal and Relational Dynamics in Multi-Session Conversationscited by 47
38 more
  1. 2024NormAd: A Framework for Measuring the Cultural Adaptability of Large Language Modelscited by 47
  2. 2024ESCoT: Towards Interpretable Emotional Support Dialogue Systemscited by 46
  3. 2023Using In-Context Learning to Improve Dialogue Safetycited by 41
  4. 2024Curry-DPO: Enhancing Alignment using Curriculum Learning & Ranked Preferencescited by 39
  5. 2024Controllable Safety Alignment: Inference-Time Adaptation to Diverse Safety Requirementscited by 35
  6. 2023COBRA Frames: Contextual Reasoning about Effects and Harms of Offensive Statementscited by 34
  7. 2023Inference-Time Policy Adapters (IPA): Tailoring Extreme-Scale LMs without Fine-tuningcited by 33
  8. 2023Mirages: On Anthropomorphism in Dialogue Systemscited by 30
  9. 2025Actions Speak Louder than Words: Agent Decisions Reveal Implicit Biases in Language Modelscited by 30
  10. 2024Culturally Aware and Adapted NLP: A Taxonomy and a Survey of the State of the Artcited by 26
  11. 2024SocialMind: LLM-based Proactive AR Social Assistive System with Human-like Perception for In-situ Live Interactionscited by 25
  12. 2023KoSBi: A Dataset for Mitigating Social Bias Risks Towards Safer Large Language Model Applicationcited by 24
  13. 2023NormDial: A Comparable Bilingual Synthetic Dialog Dataset for Modeling Social Norm Adherence and Violationcited by 24
  14. 2024Large Language Models for Manufacturingcited by 24
  15. 2024Can LLMs Recognize Toxicity? A Structured Investigation Framework and Toxicity Metriccited by 23
  16. 2024Deconstructing The Ethics of Large Language Models from Long-standing Issues to New-emerging Dilemmas: A Surveycited by 23
  17. 2023How Far Can We Extract Diverse Perspectives from Large Language Models?cited by 22
  18. 2023DialogStudio: Towards Richest and Most Diverse Unified Dataset Collection for Conversational AIcited by 21
  19. 2024NLP for Counterspeech against Hate: A Survey and How-To Guidecited by 21
  20. 2023Enabling On-Device Large Language Model Personalization with Self-Supervised Data Selection and Synthesiscited by 20
  21. 2025ReLearn: Unlearning via Learning for Large Language Modelscited by 20
  22. 2024Granular Privacy Control for Geolocation with Vision Language Modelscited by 19
  23. 2024Stress-Testing Long-Context Language Models with Lifelong ICL and Task Haystackcited by 19
  24. 2025The Sum Leaks More Than Its Parts: Compositional Privacy Risks and Mitigations in Multi-Agent Collaborationcited by 19
  25. 2024Contextualization Distillation from Large Language Model for Knowledge Graph Completioncited by 18
  26. 2022MoralDial: A Framework to Train and Evaluate Moral Dialogue Systems via Moral Discussionscited by 17
  27. 2024PolygloToxicityPrompts: Multilingual Evaluation of Neural Toxic Degeneration in Large Language Modelscited by 16
  28. 2025Cultural Learning-Based Culture Adaptation of Language Modelscited by 16
  29. 2023What Makes it Ok to Set a Fire? Iterative Self-distillation of Contexts and Rationales for Disambiguating Defeasible Social and Moral Situationscited by 15
  30. 2023Towards Safer Generative Language Models: A Survey on Safety Risks, Evaluations, and Improvementscited by 14
  31. 2023Modeling Empathic Similarity in Personal Narrativescited by 14
  32. 2023SQuARe: A Large-Scale Dataset of Sensitive Questions and Acceptable Responses Created Through Human-Machine Collaborationcited by 13
  33. 2025MIRROR: Multimodal Cognitive Reframing Therapy for Rolling with Resistancecited by 13
  34. 2023Can Language Model Moderators Improve the Health of Online Discourse?cited by 12
  35. 2023Approximating Online Human Evaluation of Social Chatbots with Promptingcited by 11
  36. 2025Crisp: Cognitive Restructuring of Negative Thoughts through Multi-turn Supportive Dialoguescited by 11
  37. 2023WHAT, WHEN, and HOW to Ground: Designing User Persona-Aware Conversational Agents for Engaging Dialoguecited by 10
  38. 2023"Oops, Did I Just Say That?" Testing and Repairing Unethical Suggestions of Large Language Models with Suggest-Critique-Reflect Processcited by 9

50 most cited of 109 All 109 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2025DialogGuard: Multi-Agent Psychosocial Safety Evaluation of Sensitive LLM Responses
  2. 2025Prosocial Behavior Detection in Player Game Chat: From Aligning Human-AI Definitions to Efficient Annotation at Scale
  3. 2026Benevolent Bias in Multi-Turn Human-Agent Dialogue
  4. 2025ProSocialAlign: Preference Conditioned Test Time Alignment in Language Models
  5. 2025SocialNLI: A Dialogue-Centric Social Inference Dataset
  6. 2024ComPeer: A Generative Conversational Agent for Proactive Peer Support
  7. 2022Improving alignment of dialogue agents via targeted human judgements
  8. 2025SI-Bench: Benchmarking Social Intelligence of Large Language Models in Human-to-Human Conversations