Language is Scary when Over-Analyzed: Unpacking Implied Misogynistic Reasoning with Argumentation Theory-Driven Prompts

This repository contains the code and resources for the project "Language is Scary when Over-Analyzed: Unpacking Implied Misogynistic Reasoning with Argumentation Theory-Driven Prompts" presented at EMNLP 2024.

Abstract

We propose misogyny detection as an Argumentative Reasoning task and we investigate the capacity of large language models (LLMs) to understand the implicit reasoning used to convey misogyny in both Italian and English. The central aim is to generate the missing reasoning link between a message and the implied meanings encoding the misogyny. Our study uses argumentation theory as a foundation to form a collection of prompts in both zero-shot and few-shot settings. These prompts integrate different techniques, including chain-of-thought reasoning and augmented knowledge. Our findings show that LLMs fall short on reasoning capabilities about misogynistic comments and that they mostly rely on their implicit knowledge derived from internalized common stereotypes about women to generate implied assumptions, rather than on inductive reasoning.

Data

Data are available upon requesting. Fill in this form to obtain them: https://docs.google.com/forms/d/1eRXilZiCwnZ7SvOVG55NJ057ReTvnH8sdWOxyDkLw7c

Contact

Arianna Muti: [email protected]

Cite

ArXiv:

@misc{muti2024languagescaryoveranalyzedunpacking,
      title={Language is Scary when Over-Analyzed: Unpacking Implied Misogynistic Reasoning with Argumentation Theory-Driven Prompts}, 
      author={Arianna Muti and Federico Ruggeri and Khalid Al-Khatib and Alberto Barrón-Cedeño and Tommaso Caselli},
      year={2024},
      eprint={2409.02519},
      archivePrefix={arXiv},
      primaryClass={cs.CL},
      url={https://arxiv.org/abs/2409.02519}, 
}

EMNLP: TBA

License

This project is licensed under CC.BY License.