Datasets, models and spaces related to hate speech detection and bias evaluation in Spanish.
María Grandury
mariagrandury
AI & ML interests
Multilingual Multicultural Responsible NLP | NLP in Spanish @ hf.co/SomosNLP
Recent Activity
updated a dataset 3 days ago
la-leaderboard/requests updated a Space 6 days ago
somosnlp-hackathon-2026/README published a Space 6 days ago
somosnlp-hackathon-2026/READMEOrganizations
-
Instruction-Following Evaluation for Large Language Models
Paper • 2311.07911 • Published • 22 -
HuggingFaceH4/mt_bench_prompts
Viewer • Updated • 80 • 8.68k • 25 -
vectara/hallucination_evaluation_model
Text Classification • Updated • 78.8k • 348 -
GAIA: a benchmark for General AI Assistants
Paper • 2311.12983 • Published • 247
Useful papers for non-English research, specially for romance languages.
-
LLaMA Beyond English: An Empirical Study on Language Capability Transfer
Paper • 2401.01055 • Published • 54 -
GlórIA -- A Generative and Open Large Language Model for Portuguese
Paper • 2402.12969 • Published -
Aurora-M: The First Open Source Multilingual Language Model Red-teamed according to the U.S. Executive Order
Paper • 2404.00399 • Published • 42
Datasets, models and spaces related to hate speech detection and bias evaluation in Spanish.
These are the resources I use and mention in my talks & workshops, for more check hf.co/ethics
-
Instruction-Following Evaluation for Large Language Models
Paper • 2311.07911 • Published • 22 -
HuggingFaceH4/mt_bench_prompts
Viewer • Updated • 80 • 8.68k • 25 -
vectara/hallucination_evaluation_model
Text Classification • Updated • 78.8k • 348 -
GAIA: a benchmark for General AI Assistants
Paper • 2311.12983 • Published • 247
Useful papers for non-English research, specially for romance languages.
-
LLaMA Beyond English: An Empirical Study on Language Capability Transfer
Paper • 2401.01055 • Published • 54 -
GlórIA -- A Generative and Open Large Language Model for Portuguese
Paper • 2402.12969 • Published -
Aurora-M: The First Open Source Multilingual Language Model Red-teamed according to the U.S. Executive Order
Paper • 2404.00399 • Published • 42