LifeTox: Unveiling Implicit Toxicity in Life Advice
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866917617786159104 |
|---|---|
| author | Kim, Minbeom Koo, Jahyun Lee, Hwanhee Park, Joonsuk Lee, Hwaran Jung, Kyomin |
| author_facet | Kim, Minbeom Koo, Jahyun Lee, Hwanhee Park, Joonsuk Lee, Hwaran Jung, Kyomin |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2311_09585 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | LifeTox: Unveiling Implicit Toxicity in Life Advice Kim, Minbeom Koo, Jahyun Lee, Hwanhee Park, Joonsuk Lee, Hwaran Jung, Kyomin Computation and Language 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. |
| title | LifeTox: Unveiling Implicit Toxicity in Life Advice |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2311.09585 |