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

• URL: http://arxiv.org/abs/2602.15757v1
• Date: Tue, 17 Feb 2026 17:45:28 GMT
• Title: Beyond Binary Classification: Detecting Fine-Grained Sexism in Social Media Videos
• Authors: Laura De Grazia, Danae Sánchez Villegas, Desmond Elliott, Mireia Farrús, Mariona Taulé,
• Abstract summary: We present FineMuSe, a new multimodal sexism detection dataset in Spanish.<n>We also introduce a comprehensive hierarchical taxonomy that encompasses forms of sexism, non-sexism, and rhetorical devices of irony and humor.<n>Our findings indicate that multimodal LLMs perform competitively with human annotators in identifying nuanced forms of sexism.
• Score: 12.430871103040275
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Online sexism appears in various forms, which makes its detection challenging. Although automated tools can enhance the identification of sexist content, they are often restricted to binary classification. Consequently, more subtle manifestations of sexism may remain undetected due to the lack of fine-grained, context-sensitive labels. To address this issue, we make the following contributions: (1) we present FineMuSe, a new multimodal sexism detection dataset in Spanish that includes both binary and fine-grained annotations; (2) we introduce a comprehensive hierarchical taxonomy that encompasses forms of sexism, non-sexism, and rhetorical devices of irony and humor; and (3) we evaluate a wide range of LLMs for both binary and fine-grained sexism detection. Our findings indicate that multimodal LLMs perform competitively with human annotators in identifying nuanced forms of sexism; however, they struggle to capture co-occurring sexist types when these are conveyed through visual cues.

Related papers

• Dataset Creation and Baseline Models for Sexism Detection in Hausa [0.0]
  This study introduces the first Hausa sexism detection dataset, developed through community engagement, qualitative coding, and data augmentation.<n>For cultural nuances and linguistic representation, we conducted a two-stage user study involving native speakers to explore how sexism is defined and articulated in everyday discourse.<n>Our findings highlight challenges in capturing cultural nuance, particularly with clarification-seeking and idiomatic expressions, and reveal a tendency for many false positives in such cases.
  arXiv  Detail & Related papers  (2025-10-30T22:57:35Z)
• Demographic Biases and Gaps in the Perception of Sexism in Large Language Models [43.77504559722899]
  We explore the capabilities of different Large Language Models to detect sexism in social media text.<n>We analyze the demographic biases present in the models and conduct a statistical analysis.<n>Our results show that, while LLMs can to some extent detect sexism when considering the overall opinion of populations, they do not accurately replicate the diversity of perceptions among different demographic groups.
  arXiv  Detail & Related papers  (2025-08-25T17:36:58Z)
• Explaining Matters: Leveraging Definitions and Semantic Expansion for Sexism Detection [9.477601265462694]
  We propose two prompt-based data augmentation techniques for sexism detection.<n>We also introduce an ensemble strategy that resolves prediction ties by aggregating complementary perspectives from multiple language models.<n>Our experimental evaluation on the EDOS dataset demonstrates state-of-the-art performance across all tasks.
  arXiv  Detail & Related papers  (2025-06-06T16:58:12Z)
• MuSeD: A Multimodal Spanish Dataset for Sexism Detection in Social Media Videos [12.555579923843641]
  We introduce a new Multimodal Spanish dataset for Sexism Detection consisting of $approx$ 11 hours of videos extracted from TikTok and BitChute.<n>We find that visual information plays a key role in labeling sexist content for both humans and models.
  arXiv  Detail & Related papers  (2025-04-15T13:16:46Z)
• The Root Shapes the Fruit: On the Persistence of Gender-Exclusive Harms in Aligned Language Models [91.86718720024825]
  We center transgender, nonbinary, and other gender-diverse identities to investigate how alignment procedures interact with pre-existing gender-diverse bias.<n>Our findings reveal that DPO-aligned models are particularly sensitive to supervised finetuning.<n>We conclude with recommendations tailored to DPO and broader alignment practices.
  arXiv  Detail & Related papers  (2024-11-06T06:50:50Z)
• Beyond Binary Gender: Evaluating Gender-Inclusive Machine Translation with Ambiguous Attitude Words [85.48043537327258]
  Existing machine translation gender bias evaluations are primarily focused on male and female genders. This study presents a benchmark AmbGIMT (Gender-Inclusive Machine Translation with Ambiguous attitude words) We propose a novel process to evaluate gender bias based on the Emotional Attitude Score (EAS), which is used to quantify ambiguous attitude words.
  arXiv  Detail & Related papers  (2024-07-23T08:13:51Z)
• A multitask learning framework for leveraging subjectivity of annotators to identify misogyny [47.175010006458436]
  We propose a multitask learning approach to enhance the performance of the misogyny identification systems. We incorporated diverse perspectives from annotators in our model design, considering gender and age across six profile groups. This research advances content moderation and highlights the importance of embracing diverse perspectives to build effective online moderation systems.
  arXiv  Detail & Related papers  (2024-06-22T15:06:08Z)
• Probing Explicit and Implicit Gender Bias through LLM Conditional Text Generation [64.79319733514266]
  Large Language Models (LLMs) can generate biased and toxic responses. We propose a conditional text generation mechanism without the need for predefined gender phrases and stereotypes.
  arXiv  Detail & Related papers  (2023-11-01T05:31:46Z)
• "I'm fully who I am": Towards Centering Transgender and Non-Binary Voices to Measure Biases in Open Language Generation [69.25368160338043]
  Transgender and non-binary (TGNB) individuals disproportionately experience discrimination and exclusion from daily life. We assess how the social reality surrounding experienced marginalization of TGNB persons contributes to and persists within Open Language Generation. We introduce TANGO, a dataset of template-based real-world text curated from a TGNB-oriented community.
  arXiv  Detail & Related papers  (2023-05-17T04:21:45Z)
• SemEval-2023 Task 10: Explainable Detection of Online Sexism [5.542286527528687]
  We introduce SemEval Task 10 on the Explainable Detection of Online Sexism (EDOS) We make three main contributions: i) a novel hierarchical taxonomy of sexist content, which includes granular vectors of sexism to aid explainability; ii) a new dataset of 20,000 social media comments with fine-grained labels, along with larger unlabelled datasets for model adaptation; andiii) baseline models as well as an analysis of the methods, results and errors for participant submissions to our task.
  arXiv  Detail & Related papers  (2023-03-07T20:28:39Z)
• "Call me sexist, but...": Revisiting Sexism Detection Using Psychological Scales and Adversarial Samples [2.029924828197095]
  We outline the different dimensions of sexism by grounding them in their implementation in psychological scales. From the scales, we derive a codebook for sexism in social media, which we use to annotate existing and novel datasets. Results indicate that current machine learning models pick up on a very narrow set of linguistic markers of sexism and do not generalize well to out-of-domain examples.
  arXiv  Detail & Related papers  (2020-04-27T13:07:46Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.