Contextualized Counterspeech: Strategies for Adaptation, Personalization, and Evaluation

AI-generated counterspeech offers a promising and scalable strat- egy to curb online toxicity through direct replies that promote civil discourse. However, current counterspeech is one-size-fits-all, lack- ing adaptation to the moderation context and the users involved. We propose and evaluate multiple strategies for generating tailored counterspeech that is adapted to the moderation context and person- alized for the moderated user. We instruct a LLaMA2-13B model to generate counterspeech, experimenting with various configurations based on different contextual information and fine-tuning strategies. We identify the configurations that generate persuasive counter- speech through a combination of quantitative indicators and human evaluations collected via a pre-registered mixed-design crowdsourc- ing experiment. Results show that contextualized counterspeech can significantly outperform state-of-the-art generic counterspeech in adequacy and persuasiveness, without compromising other char- acteristics. Our findings also reveal a poor correlation between quantitative indicators and human evaluations, suggesting that these methods assess different aspects and highlighting the need for nuanced evaluation methodologies. The effectiveness of contex- tualized AI-generated counterspeech and the divergence between human and algorithmic evaluations underscore the importance of increased human-AI collaboration in content moderation.