Automatic Construction of a Korean Toxic Instruction Dataset for Ethical Tuning of Large Language Models
Not on the map. Why?
Abstract (arXiv)
Caution: this paper may include material that could be offensive or distressing. The advent of Large Language Models (LLMs) necessitates the development of training approaches that mitigate the generation of unethical language and aptly manage toxic user queries. Given the challenges related to human labor and the scarcity of data, we present KoTox, comprising 39K unethical instruction-output pairs. This collection of automatically generated toxic instructions refines the training of LLMs and establishes a foundational framework for improving LLMs' ethical awareness and response to various toxic inputs, promoting more secure and responsible interactions in Natural Language Processing (NLP) applications.
Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Builds on 19 of its 33 references that are in the map. First posted on arXiv 30 Nov 2023.
Builds on
19 of 33 references.
References · 19
- 2020Language Models are Few-Shot Learnerscited by 32,419
- 2017Automated Hate Speech Detection and the Problem of Offensive Languagecited by 736
- 2019Predicting the Type and Target of Offensive Posts in Social Mediacited by 271
- 2021Annotators with Attitudes: How Annotator Beliefs And Identities Bias Toxic Language Detectioncited by 264
- 2022A New Generation of Perspective API: Efficient Multilingual Character-level Transformerscited by 185
- 2019Build it Break it Fix it for Dialogue Safety: Robustness from Adversarial Human Attackcited by 176
- 2020Recipes for Safety in Open-domain Chatbotscited by 170
- 2020Directions in Abusive Language Training Data: Garbage In, Garbage Outcited by 140
- 2023Explore, Establish, Exploit: Red Teaming Language Models from Scratchcited by 114
- 2021Challenges in Automated Debiasing for Toxic Language Detectioncited by 96
- 2021What Changes Can Large-scale Language Models Bring? Intensive Study on HyperCLOVA: Billions-scale Korean Generative Pretrained Transformerscited by 76
- 2020BEEP! Korean Corpus of Online News Comments for Toxic Speech Detectioncited by 39
7 more
- 2023KoSBi: A Dataset for Mitigating Social Bias Risks Towards Safer Large Language Model Applicationcited by 24
- 2023A Technical Report for Polyglot-Ko: Open-Source Large-Scale Korean Language Modelscited by 22
- 2021Data Expansion using Back Translation and Paraphrasing for Hate Speech Detectioncited by 20
- 2022Reward Modeling for Mitigating Toxicity in Transformer-based Language Modelscited by 18
- 2023Adding Instructions during Pretraining: Effective Way of Controlling Toxicity in Language Modelscited by 16
- 2023An Invariant Learning Characterization of Controlled Text Generationcited by 9
- 2022APEACH: Attacking Pejorative Expressions with Analysis on Crowd-Generated Hate Speech Evaluation Datasetscited by 8
Built on by
Citing papers per year: 2025: 1.
1 paper.