Zobrazit v EDS

Revisiting Implicitly Abusive Language Detection: Evaluating LLMs in Zero-Shot and Few-Shot Settings

Uloženo v:
Podrobná bibliografie
Název: Revisiting Implicitly Abusive Language Detection: Evaluating LLMs in Zero-Shot and Few-Shot Settings
Autoři: Jaremko, Julia, Gromann, Dagmar, Wiegand, Michael
Informace o vydavateli: 2025.
Rok vydání: 2025
Témata: 602011 Computerlinguistik, 602011 Computational linguistics
Popis: Implicitly abusive language (IAL), unlike its explicit counterpart, lacks overt slurs or unambiguously offensive keywords, such as “bimbo” or “scum”, making it challenging to detect and mitigate. While current research predominantly focuses on explicitly abusive language, the subtler and more covert forms of IAL remain insufficiently studied. The rapid advancement and widespread adoption of large language models (LLMs) have opened new possibilities for various NLP tasks, but their application to IAL detection has been limited. We revisit three very recent challenging datasets of IAL and investigate the potential of LLMs to enhance the detection of IAL in English through zero-shot and few-shot prompting approaches. We evaluate the models’ capabilities in classifying sentences directly as either IAL or benign, and in extracting linguistic features associated with IAL. Our results indicate that classifiers trained on features extracted by advanced LLMs outperform the best previously reported results, achieving near-human performance.
Druh dokumentu: Conference object
Jazyk: English
Přístupová URL adresa: https://ucrisportal.univie.ac.at/de/publications/bb883f2d-d6a2-4831-abaf-fde2ddccd8bd
Přístupové číslo: edsair.dris...00911..343fd4d7c12b633a204a740767a8501b
Databáze: OpenAIRE
Popis
Abstrakt:Implicitly abusive language (IAL), unlike its explicit counterpart, lacks overt slurs or unambiguously offensive keywords, such as “bimbo” or “scum”, making it challenging to detect and mitigate. While current research predominantly focuses on explicitly abusive language, the subtler and more covert forms of IAL remain insufficiently studied. The rapid advancement and widespread adoption of large language models (LLMs) have opened new possibilities for various NLP tasks, but their application to IAL detection has been limited. We revisit three very recent challenging datasets of IAL and investigate the potential of LLMs to enhance the detection of IAL in English through zero-shot and few-shot prompting approaches. We evaluate the models’ capabilities in classifying sentences directly as either IAL or benign, and in extracting linguistic features associated with IAL. Our results indicate that classifiers trained on features extracted by advanced LLMs outperform the best previously reported results, achieving near-human performance.