LifeTox: Unveiling Implicit Toxicity in Life Advice
Preprint
Abstract
As large language models become increasingly integrated into daily life, detecting implicit toxicity across diverse contexts is crucial. To this end, we introduce LifeTox, a dataset designed for identifying implicit toxicity within a broad range of advice-seeking scenarios. Unlike existing safety datasets, LifeTox comprises diverse contexts derived from personal experiences through open-ended questions. Experiments demonstrate that RoBERTa fine-tuned on LifeTox matches or surpasses the zero-shot performance of large language models in toxicity classification tasks. These results underscore the efficacy of LifeTox in addressing the complex challenges inherent in implicit toxicity. We open-sourced the dataset\footnote{\url{https://huggingface.co/datasets/mbkim/LifeTox}} and the LifeTox moderator family; 350M, 7B, and 13B.Keywords
All Related Versions
This publication has 19 references indexed in Scilit:
- Bias and Fairness in Large Language Models: A SurveyComputational Linguistics, 2024
- Handling Bias in Toxic Speech Detection: A SurveyACM Computing Surveys, 2023
- ChatGPT for good? On opportunities and challenges of large language models for educationLearning and Individual Differences, 2023
- Influence of Rapport and Social Presence with an AI Psychotherapy Chatbot on Users’ Self-DisclosureInternational Journal of Human–Computer Interaction, 2022
- Latent Hatred: A Benchmark for Understanding Implicit Hate Speech2021
- StereoSet: Measuring stereotypical bias in pretrained language models2021
- Implicitly Abusive Language – What does it actually look like and why are we not getting there?2021
- Hurdles to Progress in Long-form Question Answering2021
- Toxicity Detection: Does Context Really Matter?2020
- Social Bias Frames: Reasoning about Social and Power Implications of Language2020