8 papers
xList-Hate: A Checklist-Based Framework for Interpretable and Generalizable Hate Speech Detection
Adrián Girón, Pablo Miralles, Javier Huertas-Tato +2
Hate speech detection is commonly framed as a direct binary classification problem despite being a composite concept defined through multiple interacting factors that vary across l…
Supervised Contrastive Learning for Few-Shot AI-Generated Image Detection and Attribution
Jaime Álvarez Urueña, David Camacho, Javier Huertas Tato
The rapid advancement of generative artificial intelligence has enabled the creation of synthetic images that are increasingly indistinguishable from authentic content, posing sign…
One-shot Style Transfer LLM log-probabilities for Authorship Attribution and Verification
Pablo Miralles-González, Javier Huertas-Tato, Alejandro Martín +1
Computational stylometry studies writing style through quantitative textual patterns, enabling applications such as authorship attribution, identity linking, and plagiarism detecti…
Evaluating the Scalability and Adversarial Generalization of GRPO-Trained NLI Models
Pablo Miralles-González, Javier Huertas-Tato, Alejandro Martín +1
Natural Language Inference (NLI) is a central task in natural language understanding with applications in fact-checking, question answering, and information retrieval. Despite its…
On the locality bias and results in the Long Range Arena
Pablo Miralles-González, Javier Huertas-Tato, Alejandro Martín +1
The Long Range Arena (LRA) benchmark was designed to evaluate the performance of Transformer improvements and alternatives in long-range dependency modeling tasks. The Transformer…
Not all tokens are created equal: Perplexity Attention Weighted Networks for AI generated text detection
Pablo Miralles-González, Javier Huertas-Tato, Alejandro Martín +1
The rapid advancement in large language models (LLMs) has significantly enhanced their ability to generate coherent and contextually relevant text, raising concerns about the misus…
Isolating authorship from content with semantic embeddings and contrastive learning
Javier Huertas-Tato, Adrián Girón-Jiménez, Alejandro Martín +1
Authorship has entangled style and content inside. Authors frequently write about the same topics in the same style, so when different authors write about the exact same topic the…
A CLIP-based siamese approach for meme classification
Javier Huertas-Tato, Christos Koutlis, Symeon Papadopoulos +2
Memes are an increasingly prevalent element of online discourse in social networks, especially among young audiences. They carry ideas and messages that range from humorous to hate…