LLM generated responses to mitigate the impact of hate speech
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2023
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866918042117603328 |
|---|---|
| author | Podolak, Jakub Łukasik, Szymon Balawender, Paweł Ossowski, Jan Piotrowski, Jan Bąkowicz, Katarzyna Sankowski, Piotr |
| author_facet | Podolak, Jakub Łukasik, Szymon Balawender, Paweł Ossowski, Jan Piotrowski, Jan Bąkowicz, Katarzyna Sankowski, Piotr |
| contents | In this study, we explore the use of Large Language Models (LLMs) to counteract hate speech. We conducted the first real-life A/B test assessing the effectiveness of LLM-generated counter-speech. During the experiment, we posted 753 automatically generated responses aimed at reducing user engagement under tweets that contained hate speech toward Ukrainian refugees in Poland.
Our work shows that interventions with LLM-generated responses significantly decrease user engagement, particularly for original tweets with at least ten views, reducing it by over 20%. This paper outlines the design of our automatic moderation system, proposes a simple metric for measuring user engagement and details the methodology of conducting such an experiment. We discuss the ethical considerations and challenges in deploying generative AI for discourse moderation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2311_16905 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | LLM generated responses to mitigate the impact of hate speech Podolak, Jakub Łukasik, Szymon Balawender, Paweł Ossowski, Jan Piotrowski, Jan Bąkowicz, Katarzyna Sankowski, Piotr Computers and Society I.2.7 In this study, we explore the use of Large Language Models (LLMs) to counteract hate speech. We conducted the first real-life A/B test assessing the effectiveness of LLM-generated counter-speech. During the experiment, we posted 753 automatically generated responses aimed at reducing user engagement under tweets that contained hate speech toward Ukrainian refugees in Poland. Our work shows that interventions with LLM-generated responses significantly decrease user engagement, particularly for original tweets with at least ten views, reducing it by over 20%. This paper outlines the design of our automatic moderation system, proposes a simple metric for measuring user engagement and details the methodology of conducting such an experiment. We discuss the ethical considerations and challenges in deploying generative AI for discourse moderation. |
| title | LLM generated responses to mitigate the impact of hate speech |
| topic | Computers and Society I.2.7 |
| url | https://arxiv.org/abs/2311.16905 |