Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

AIWizards at MULTIPRIDE: A Hierarchical Approach to Slur Reclamation Detection

Luca Tedeschini, Matteo Fasulo

TL;DR

This work addresses Subtask B of the MultiPRIDE shared task at EVALITA 2026 by proposing a hierarchical approach to modeling the slur reclamation process, and argues that more fine-grained hierarchical modeling of user identity and discourse context may further improve the detection of reclaimed language.

Abstract

Detecting reclaimed slurs represents a fundamental challenge for hate speech detection systems, as the same lexcal items can function either as abusive expressions or as in-group affirmations depending on social identity and context. In this work, we address Subtask B of the MultiPRIDE shared task at EVALITA 2026 by proposing a hierarchical approach to modeling the slur reclamation process. Our core assumption is that members of the LGBTQ+ community are more likely, on average, to employ certain slurs in a eclamatory manner. Based on this hypothesis, we decompose the task into two stages. First, using a weakly supervised LLM-based annotation, we assign fuzzy labels to users indicating the likelihood of belonging to the LGBTQ+ community, inferred from the tweet and the user bio. These soft labels are then used to train a BERT-like model to predict community membership, encouraging the model to learn latent representations associated with LGBTQ+ identity. In the second stage, we integrate this latent space with a newly initialized model for the downstream slur reclamation detection task. The intuition is that the first model encodes user-oriented sociolinguistic signals, which are then fused with representations learned by a model pretrained for hate speech detection. Experimental results on Italian and Spanish show that our approach achieves performance statistically comparable to a strong BERT-based baseline, while providing a modular and extensible framework for incorporating sociolinguistic context into hate speech modeling. We argue that more fine-grained hierarchical modeling of user identity and discourse context may further improve the detection of reclaimed language. We release our code at https://github.com/LucaTedeschini/multipride.

AIWizards at MULTIPRIDE: A Hierarchical Approach to Slur Reclamation Detection

TL;DR

This work addresses Subtask B of the MultiPRIDE shared task at EVALITA 2026 by proposing a hierarchical approach to modeling the slur reclamation process, and argues that more fine-grained hierarchical modeling of user identity and discourse context may further improve the detection of reclaimed language.

Abstract

Detecting reclaimed slurs represents a fundamental challenge for hate speech detection systems, as the same lexcal items can function either as abusive expressions or as in-group affirmations depending on social identity and context. In this work, we address Subtask B of the MultiPRIDE shared task at EVALITA 2026 by proposing a hierarchical approach to modeling the slur reclamation process. Our core assumption is that members of the LGBTQ+ community are more likely, on average, to employ certain slurs in a eclamatory manner. Based on this hypothesis, we decompose the task into two stages. First, using a weakly supervised LLM-based annotation, we assign fuzzy labels to users indicating the likelihood of belonging to the LGBTQ+ community, inferred from the tweet and the user bio. These soft labels are then used to train a BERT-like model to predict community membership, encouraging the model to learn latent representations associated with LGBTQ+ identity. In the second stage, we integrate this latent space with a newly initialized model for the downstream slur reclamation detection task. The intuition is that the first model encodes user-oriented sociolinguistic signals, which are then fused with representations learned by a model pretrained for hate speech detection. Experimental results on Italian and Spanish show that our approach achieves performance statistically comparable to a strong BERT-based baseline, while providing a modular and extensible framework for incorporating sociolinguistic context into hate speech modeling. We argue that more fine-grained hierarchical modeling of user identity and discourse context may further improve the detection of reclaimed language. We release our code at https://github.com/LucaTedeschini/multipride.
Paper Structure (23 sections, 2 equations, 1 figure, 7 tables)

This paper contains 23 sections, 2 equations, 1 figure, 7 tables.

Table of Contents

  1. Introduction
  2. Task and Data
  3. Task Definition
  4. Data Characteristics
  5. Why User Content Matters
  6. LLM-Assisted Proxy Labeling for the User Encoder
  7. System Description
  8. Baseline Model
  9. Dual-Encoder Architecture
  10. User Encoder
  11. Training Recipe
  12. Hyperparameters
  13. Loss Function
  14. Feature Fusion via Learned Gating
  15. Training Procedure
  16. ...and 8 more sections

Figures (1)

  • Figure 1: Overview of the proposed dual-encoder architecture. A text encoder and a user encoder process the concatenation of tweet and user biography in parallel. Their final hidden representations are fused through a learned gating mechanism to produce a user-aware representation for slur reclamation classification.