mlsanityData to 14 Oct 2026

SQuARe: A Large-Scale Dataset of Sensitive Questions and Acceptable Responses Created Through Human-Machine Collaboration

arXiv 2305.17696 ↗First posted 28 May 2023cs.CLcited by 13

Not on the map. Why?

Abstract (arXiv)

The potential social harms that large language models pose, such as generating offensive content and reinforcing biases, are steeply rising. Existing works focus on coping with this concern while interacting with ill-intentioned users, such as those who explicitly make hate speech or elicit harmful responses. However, discussions on sensitive issues can become toxic even if the users are well-intentioned. For safer models in such scenarios, we present the Sensitive Questions and Acceptable Response (SQuARe) dataset, a large-scale Korean dataset of 49k sensitive questions with 42k acceptable and 46k non-acceptable responses. The dataset was constructed leveraging HyperCLOVA in a human-in-the-loop manner based on real news headlines. Experiments show that acceptable response generation significantly improves for HyperCLOVA and GPT-3, demonstrating the efficacy of this dataset.

Key facts. Cited by 13 papers in the map (data through 14 Oct 2026). Builds on 36 of its 43 references that are in the map. First posted on arXiv 28 May 2023.

Builds on

36 of 43 references.

References · 36

  1. 2020Language Models are Few-Shot Learnerscited by 32,419
  2. 2022Training language models to follow instructions with human feedbackcited by 15,971
  3. 2021On the Opportunities and Risks of Foundation Modelscited by 4,033
  4. 2022Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedbackcited by 3,720
  5. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  6. 2020Making Pre-trained Language Models Better Few-shot Learnerscited by 1,256
  7. 2020RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Modelscited by 1,255
  8. 2022Red Teaming Language Models with Language Modelscited by 1,026
  9. 2021Ethical and social risks of harm from Language Modelscited by 966
  10. 2020StereoSet: Measuring stereotypical bias in pretrained language modelscited by 924
  11. 2017Automated Hate Speech Detection and the Problem of Offensive Languagecited by 736
  12. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 729
24 more
  1. 2020Aligning AI With Shared Human Valuescited by 723
  2. 2019Universal Adversarial Triggers for Attacking and Analyzing NLPcited by 722
  3. 2021Cross-Task Generalization via Natural Language Crowdsourcing Instructionscited by 699
  4. 2020Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamicscited by 425
  5. 2022ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detectioncited by 422
  6. 2019Social Bias Frames: Reasoning about Social and Power Implications of Languagecited by 380
  7. 2020Social Chemistry 101: Learning to Reason about Social and Moral Normscited by 255
  8. 2021Societal Biases in Language Generation: Progress and Challengescited by 202
  9. 2022WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creationcited by 202
  10. 2021GPT3Mix: Leveraging Large-scale Language Models for Text Augmentationcited by 187
  11. 2022A New Generation of Perspective API: Efficient Multilingual Character-level Transformerscited by 185
  12. 2021Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasetscited by 181
  13. 2021Challenges in Detoxifying Language Modelscited by 169
  14. 2021Alignment of Language Agentscited by 168
  15. 2022The 'Problem' of Human Label Variation: On Ground Truth in Data, Modeling and Evaluationcited by 139
  16. 2020Moral Stories: Situated Reasoning about Norms, Intents, Actions, and their Consequencescited by 135
  17. 2021Reframing Human-AI Collaboration for Generating Free-Text Explanationscited by 125
  18. 2022ProsocialDialog: A Prosocial Backbone for Conversational Agentscited by 109
  19. 2020Scruples: A Corpus of Community Ethical Judgments on 32,000 Real-Life Anecdotescited by 101
  20. 2021What Changes Can Large-scale Language Models Bring? Intensive Study on HyperCLOVA: Billions-scale Korean Generative Pretrained Transformerscited by 76
  21. 2021Neural Data Augmentation via Example Extrapolationcited by 54
  22. 2020SOLID: A Large-Scale Semi-Supervised Dataset for Offensive Language Identificationcited by 53
  23. 2022Building a Role Specified Open-Domain Dialogue System Leveraging Large-Scale Language Modelscited by 26
  24. 2022KOLD: Korean Offensive Language Datasetcited by 14

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2023: 1; 2024: 6; 2025: 5; 2026: 1.

13 papers.

Sort:

By area:

  1. 2024Can LLMs Recognize Toxicity? A Structured Investigation Framework and Toxicity Metriccited by 23
  2. 2023LifeTox: Unveiling Implicit Toxicity in Life Advicecited by 14
  3. 2025Guardians and Offenders: A Survey on Harmful Content Generation and Safety Mitigation of LLMcited by 9
  4. 2024SLM as Guardian: Pioneering AI Safety with Small Language Modelscited by 7
  5. 2024PSYDIAL: Personality-based Synthetic Dialogue Generation using Large Language Modelscited by 6
  6. 2025From KMMLU-Redux to KMMLU-Pro: A Professional Korean Benchmark Suite for LLM Evaluationcited by 6
  7. 2025Multi-Step Reasoning in Korean and the Emergent Miragecited by 4
  8. 2026CAGE: A Framework for Culturally Adaptive Red-Teaming Benchmark Generationcited by 3
  9. 2024On the Consideration of AI Openness: Can Good Intent Be Abused?cited by 2
  10. 2024Representing the Under-Represented: Cultural and Core Capability Benchmarks for Developing Thai Large Language Modelscited by 2
  11. 2024Taxonomy and Analysis of Sensitive User Queries in Generative AI Searchcited by 1
  12. 2025"There Is No Such Thing as a Dumb Question," But There Are Good Onescited by 0
1 more
  1. 2025Delving into Multilingual Ethical Bias: The MSQAD with Statistical Hypothesis Tests for Large Language Modelscited by 0