# KoSBi: A Dataset for Mitigating Social Bias Risks Towards Safer Large Language Model Application

arXiv: https://arxiv.org/abs/2305.17701 · HTML: https://mlsanity.com/paper/2305.17701/

## Key facts

- Cited by 24 papers in the map (data through 14 Oct 2026).
- Builds on 40 of its 52 references that are in the map.
- First posted on arXiv 28 May 2023.

## Builds on (most cited first, 30 of 40)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training a Helpful and Harmless Assistant with Reinforcement Learning from Human Feedback](https://mlsanity.com/paper/2204.05862/) (2022)
- [Making Pre-trained Language Models Better Few-shot Learners](https://mlsanity.com/paper/2012.15723/) (2020)
- [RealToxicityPrompts: Evaluating Neural Toxic Degeneration in Language Models](https://mlsanity.com/paper/2009.11462/) (2020)
- [Red Teaming Language Models with Language Models](https://mlsanity.com/paper/2202.03286/) (2022)
- [Language (Technology) is Power: A Critical Survey of "Bias" in NLP](https://mlsanity.com/paper/2005.14050/) (2020)
- [StereoSet: Measuring stereotypical bias in pretrained language models](https://mlsanity.com/paper/2004.09456/) (2020)
- [Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learned](https://mlsanity.com/paper/2209.07858/) (2022)
- [Aligning AI With Shared Human Values](https://mlsanity.com/paper/2008.02275/) (2020)
- [Universal Adversarial Triggers for Attacking and Analyzing NLP](https://mlsanity.com/paper/1908.07125/) (2019)
- [CrowS-Pairs: A Challenge Dataset for Measuring Social Biases in Masked Language Models](https://mlsanity.com/paper/2010.00133/) (2020)
- [BBQ: A Hand-Built Bias Benchmark for Question Answering](https://mlsanity.com/paper/2110.08193/) (2021)
- [Cross-Task Generalization via Natural Language Crowdsourcing Instructions](https://mlsanity.com/paper/2104.08773/) (2021)
- [Gender Bias in Coreference Resolution: Evaluation and Debiasing Methods](https://mlsanity.com/paper/1804.06876/) (2018)
- [The Woman Worked as a Babysitter: On Biases in Language Generation](https://mlsanity.com/paper/1909.01326/) (2019)
- [Gender Bias in Coreference Resolution](https://mlsanity.com/paper/1804.09301/) (2018)
- [Dataset Cartography: Mapping and Diagnosing Datasets with Training Dynamics](https://mlsanity.com/paper/2009.10795/) (2020)
- [ToxiGen: A Large-Scale Machine-Generated Dataset for Adversarial and Implicit Hate Speech Detection](https://mlsanity.com/paper/2203.09509/) (2022)
- [Social Bias Frames: Reasoning about Social and Power Implications of Language](https://mlsanity.com/paper/1911.03891/) (2019)
- [Social Chemistry 101: Learning to Reason about Social and Moral Norms](https://mlsanity.com/paper/2011.00620/) (2020)
- [HateCheck: Functional Tests for Hate Speech Detection Models](https://mlsanity.com/paper/2012.15606/) (2020)
- [WANLI: Worker and AI Collaboration for Natural Language Inference Dataset Creation](https://mlsanity.com/paper/2201.05955/) (2022)
- [Process for Adapting Language Models to Society (PALMS) with Values-Targeted Datasets](https://mlsanity.com/paper/2106.10328/) (2021)
- [Alignment of Language Agents](https://mlsanity.com/paper/2103.14659/) (2021)
- [Moral Stories: Situated Reasoning about Norms, Intents, Actions, and their Consequences](https://mlsanity.com/paper/2012.15738/) (2020)
- [UnQovering Stereotyping Biases via Underspecified Questions](https://mlsanity.com/paper/2010.02428/) (2020)
- [KLUE: Korean Language Understanding Evaluation](https://mlsanity.com/paper/2105.09680/) (2021)
- [ProsocialDialog: A Prosocial Backbone for Conversational Agents](https://mlsanity.com/paper/2205.12688/) (2022)
- [Detoxifying Language Models Risks Marginalizing Minority Voices](https://mlsanity.com/paper/2104.06390/) (2021)
- [Scruples: A Corpus of Community Ethical Judgments on 32,000 Real-Life Anecdotes](https://mlsanity.com/paper/2008.09094/) (2020)

## Built on by (most cited first, 24 of 24)

- [Towards Measuring and Modeling "Culture" in LLMs: A Survey](https://mlsanity.com/paper/2403.15412/) (2024)
- [KoBBQ: Korean Bias Benchmark for Question Answering](https://mlsanity.com/paper/2307.16778/) (2023)
- [Bias Testing and Mitigation in LLM-based Code Generation](https://mlsanity.com/paper/2309.14345/) (2023)
- [Prompt Packer: Deceiving LLMs through Compositional Instruction with Hidden Attacks](https://mlsanity.com/paper/2310.10077/) (2023)
- [Navigating LLM Ethics: Advancements, Challenges, and Future Directions](https://mlsanity.com/paper/2406.18841/) (2024)
- [Culturally Aware and Adapted NLP: A Taxonomy and a Survey of the State of the Art](https://mlsanity.com/paper/2406.03930/) (2024)
- [PromptKD: Distilling Student-Friendly Knowledge for Generative Language Models via Prompt Tuning](https://mlsanity.com/paper/2402.12842/) (2024)
- [Optimizing Language Augmentation for Multilingual Large Language Models: A Case Study on Korean](https://mlsanity.com/paper/2403.10882/) (2024)
- [LLMs with Industrial Lens: Deciphering the Challenges and Prospects -- A Survey](https://mlsanity.com/paper/2402.14558/) (2024)
- [AdvisorQA: Towards Helpful and Harmless Advice-seeking Question Answering with Collective Intelligence](https://mlsanity.com/paper/2404.11826/) (2024)
- [A Review of Incorporating Psychological Theories in LLMs](https://mlsanity.com/paper/2505.00003/) (2025)
- [No Size Fits All: The Perils and Pitfalls of Leveraging LLMs Vary with Company Size](https://mlsanity.com/paper/2408.01444/) (2024)
- [Benchmarks Are Not Atomic: Composition-Aware LLM Evaluation using BenchHub](https://mlsanity.com/paper/2506.00482/) (2025)
- [CAGE: A Framework for Culturally Adaptive Red-Teaming Benchmark Generation](https://mlsanity.com/paper/2602.20170/) (2026)
- [On the Consideration of AI Openness: Can Good Intent Be Abused?](https://mlsanity.com/paper/2403.06537/) (2024)
- [Social Bias in Multilingual Language Models: A Survey](https://mlsanity.com/paper/2508.20201/) (2025)
- [OccuQuest: Mitigating Occupational Bias for Inclusive Large Language Models](https://mlsanity.com/paper/2310.16517/) (2023)
- [Automatic Construction of a Korean Toxic Instruction Dataset for Ethical Tuning of Large Language Models](https://mlsanity.com/paper/2311.18215/) (2023)
- [Reducing Large Language Model Bias with Emphasis on 'Restricted Industries': Automated Dataset Augmentation and Prejudice Quantification](https://mlsanity.com/paper/2403.13925/) (2024)
- [The Dual-use Dilemma in LLMs: Do Empowering Ethical Capacities Make a Degraded Utility?](https://mlsanity.com/paper/2501.13952/) (2025)
- [From National Curricula to Cultural Awareness: Constructing Open-Ended Culture-Specific Question Answering Dataset](https://mlsanity.com/paper/2601.04632/) (2026)
- [Marginal Alignment Does Not Guarantee Joint-Distribution Fidelity: An Official-Reference Audit of Nemotron-Personas-Korea with Cross-Locale Replication](https://mlsanity.com/paper/2606.12433/) (2026)
- [Off-Target Effects of Response-Style Alignment in a Korean 27B Language Model](https://mlsanity.com/paper/2609.11291/) (2026)
- [CultureMINE: Datasets and Methods for Improving the Cultural Capabilities of NLP Systems](https://mlsanity.com/paper/2609.22494/) (2026)

## Abstract (arXiv)

Large language models (LLMs) learn not only natural text generation abilities but also social biases against different demographic groups from real-world data. This poses a critical risk when deploying LLM-based applications. Existing research and resources are not readily applicable in South Korea due to the differences in language and culture, both of which significantly affect the biases and targeted demographic groups. This limitation requires localized social bias datasets to ensure the safe and effective deployment of LLMs. To this end, we present KO SB I, a new social bias dataset of 34k pairs of contexts and sentences in Korean covering 72 demographic groups in 15 categories. We find that through filtering-based moderation, social biases in generated content can be reduced by 16.47%p on average for HyperCLOVA (30B and 82B), and GPT-3.