Multilingual large language model: A survey of resources, taxonomy and frontiers
Multilingual Large Language Models are capable of using powerful Large Language
Models to handle and respond to queries in multiple languages, which achieves remarkable …
Models to handle and respond to queries in multiple languages, which achieves remarkable …
Demonstrations are all you need: Advancing offensive content paraphrasing using in-context learning
Paraphrasing of offensive content is a better alternative to content removal and helps
improve civility in a communication environment. Supervised paraphrasers; however, rely …
improve civility in a communication environment. Supervised paraphrasers; however, rely …
[PDF][PDF] Discgen: A framework for discourse-informed counterspeech generation
S Hassan, M Alikhani - Proceedings of the 13th International Joint …, 2023 - aclanthology.org
Counterspeech can be an effective method for battling hateful content on social media.
Automated counterspeech generation can aid in this process. Generated counterspeech …
Automated counterspeech generation can aid in this process. Generated counterspeech …
LLM-Mod: Can Large Language Models Assist Content Moderation?
M Kolla, S Salunkhe, E Chandrasekharan… - Extended Abstracts of the …, 2024 - dl.acm.org
Content moderation is critical for maintaining healthy online spaces. However, it remains a
predominantly manual task. Moderators are often exhausted by low moderator-to-posts ratio …
predominantly manual task. Moderators are often exhausted by low moderator-to-posts ratio …
D-CALM: A dynamic clustering-based active learning approach for mitigating bias
S Hassan, M Alikhani - arXiv preprint arXiv:2305.17013, 2023 - arxiv.org
Despite recent advancements, NLP models continue to be vulnerable to bias. This bias often
originates from the uneven distribution of real-world data and can propagate through the …
originates from the uneven distribution of real-world data and can propagate through the …
Promoting Constructive Deliberation: Reframing for Receptiveness
To promote constructive discussion of controversial topics online, we propose automatic
reframing of disagreeing responses to signal receptiveness while preserving meaning …
reframing of disagreeing responses to signal receptiveness while preserving meaning …
Like a Good Nearest Neighbor: Practical Content Moderation and Text Classification
L Bates, I Gurevych - arXiv preprint arXiv:2302.08957, 2023 - arxiv.org
Few-shot text classification systems have impressive capabilities but are infeasible to deploy
and use reliably due to their dependence on prompting and billion-parameter language …
and use reliably due to their dependence on prompting and billion-parameter language …
Language identification for typologicallysimilar low-resource languages: A case study of Meänkieli, Kven and Finnish
J Larsson - 2024 - diva-portal.org
This studies examines different methods of language identification for the languages
Meänkieli, Kven and Finnish. The methods explored are two n-gram based classifiers; Naive …
Meänkieli, Kven and Finnish. The methods explored are two n-gram based classifiers; Naive …