Synapse
⌘+K
Synapse
PulseExploreClubsResearchersJournals
Instagram
HomeClubsExplore
July 26, 2026Computational LinguisticsOpen Access

Beyond Binary Classification: Detecting Fine-Grained Sexism in Social Media Videos

View Full Paper
Ask AI
Bookmark
Share

Authors

LGLaura De GraziaDVDanae Sánchez VillegasDEDesmond Elliott

Discussion

Loading...

Member takes

Overview

Randomized trial evaluates nuanced sexism detection in videos, suggesting multimodal models rival human annotators.

Key Points

  • The aim is to enhance the detection of subtle sexism in social media videos through fine-grained classification.
  • Developed FineMuSe dataset with binary and fine-grained annotations in Spanish.
  • Created a hierarchical taxonomy for categorizing different forms of sexism and related rhetorical devices.
  • Evaluated multiple large language models for their performance in detecting nuanced sexism.
  • Multimodal LLMs achieved competitive accuracy with human annotators in detecting nuanced sexism.
  • The models showed limitations in interpreting visual cues associated with sexist content.

Cite This Study

Grazia et al. (2026) studied this question.

synapsesocial.com/papers/6a65a468d3aea3239cd76fdfhttps://doi.org/10.1162/coli.a.646
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1Bilingual Sexism Classification: Fine-Tuned XLM-RoBERTa and GPT-3.5 Few-Shot Learning2024
  2. 2MIMIC: Misogyny Identification in Multimodal Internet Content in Hindi-English Code-Mixed Language2024 · 27 citations
  3. 3Unveiling Misogyny Memes: A Multimodal Analysis of Modality Effects on Identification2024 · 3 citations
  4. 4Automatic Detection of Multilingual Misogynistic Content in Social Media Data Based on Machine Learning Approach2024 · 2 citations
  5. 5Evaluating Online Sexism Detection: A Comparative Study of Machine Learning Models using the EDOS Dataset2024 · 3 citations