Subtask 1: Misogyny Detection (binary classification): This task aims to classify lyrics containing misogynistic discourse at the song level. It is a binary classification task:

• Misogynistic (M): contains hate speech or contempt directed at women or perpetuates harmful gender stereotypes that promote the subordination or objectification of women.

• Non-misogynistic (NM): the lyrics do not include hateful or derogatory language toward women. They may address issues related to women without perpetuating stereotypes or negative attitudes.

Subtask 2: Misogyny Type Classification (multilabel): This task aims to predict the types of misogynistic discourse present in the lyrics of a song. The categories are as follows:

• Sexualization (S): phrases that describe or suggest sexual acts, sexual language, or innuendo.

• Violence (V): phrases that refer to physical or verbal aggression, threats, or violent actions.

• Hate (H): phrases containing offensive or discriminatory language, expressions of contempt or hostility toward a group or individual.

• Not related (NR): phrases that do not fall into the above categories and lack sexual, violent, or hateful content.

Subtask 3: Stereotype Identification (binary classification): This task aims to classify phrases from song lyrics that contain gender-based stereotypes. It is a binary classification task:

• Stereotypical (S): Lyrics depict or reinforce generalized, oversimplified, or biased beliefs about gender roles, traits, or behaviors. These stereotypes may promote assumptions about how women or men should behave, their capabilities, social roles, or emotional characteristics (e.g., women as submissive or emotional; men as dominant or unemotional).
  
  

• Non-Stereotypical (NS): Lyrics do not include generalized or biased beliefs about gender. They may mention men or women, or discuss gender-related themes, but without reinforcing restrictive roles, assumptions, or attributes linked to a specific gender.