Dataset

Untangling Hate Speech Definitions: A Semantic Componential Analysis Across Cultures and Domains
HateDefCon breaks 493 hate speech definitions from five domains into semantic components and shows LLM hate speech detection changes with the definition in the prompt.
Katerina Korre, Arianna Muti, Federico Ruggeri, Alberto Barrón-Cedeño
Cite DOI URL
Untangling Hate Speech Definitions: A Semantic Componential Analysis Across Cultures and Domains
A Dataset of Argumentative Dialogues on Scientific Papers
ArgSciChat collects 41 dialogues in which scientists question and argue about NLP papers, a hard test for dialogue agents that must explain and defend answers.
Federico Ruggeri, Mohsen Mesgar, Iryna Gurevych
Cite DOI URL
Inception Models for Fashion Image Captioning: An Extensive Study on Multiple Datasets
Fashion-Cap, a new dataset for writing captions of fashion product photos, tested with InceptionV3 captioners against two existing datasets of different difficulty.
Mirko Del Moro, Serban Cristian Tudosie, Francesco Vannoni, Andrea Galassi, Federico Ruggeri
Cite DOI URL