Corpora

MCIF

MCIF (Multimodal Crosslingual Instruction Following) is a multilingual human-annotated benchmark based on scientific talks that is designed to evaluate instruction-following in crosslingual, multimodal settings over both short-...

Read More

mGeNTE

mGeNTE (Multilingual Gender-Neutral Translation Evaluation) is a natural, multilingual corpus designed to benchmark gender-neutral language and automatic translation.mGente is built upon European Parliament speech data extracted...

Read More

MOSEL

The MOSEL corpus is a multilingual dataset collection including up to 950K hours of open-source speech recordings covering the 24 official languages of the European Union. We collect data by surveying labeled and unlabeled...

Read More

Speech-MASSIVE

Spoken Language Understanding (SLU) involves interpreting spoken input using Natural Language Processing (NLP). Voice assistants like Alexa and Siri are real-world examples of SLU applications. The core tasks in SLU include...

Read More

INES

The INclusive Evaluation Suite (INES) is a test set designed to assess MT systems ability to produce gender-inclusive translations for the German→English language pair. By design, each German source sentence in INES includes an...

Read More

GeNTE

GeNTE (Gender-Neutral Translation Evaluation) is a natural, bilingual corpus designed to benchmark the ability of machine translation systems to generate gender-neutral translations. Built from European Parliament speeches,...

Read More
Loading

⏰ 4 days left to apply!
πŸŽ“ 2 PhD positions still open:
🎯 Human-Centred Evaluation Frameworks for Multilingual Technologies
✨ Multimedia Personalization with Multimodal Large Language Models
πŸ“… Deadline: 15 May 2026
πŸ”— Full details: https://iecs.unitn.it/education/admission/call-for-application

🌍 @lina_conti and @luisabentivogli are heading to #LREC2026 in Palma! They'll present two papers:
πŸ“„ "Voice, Bias, and Coreference: An Interpretability Study of Gender in Speech Translation"
Paper link:

πŸ€” What Matters in Data for DPO? I asked myself this question a few days ago while trying to understand how to generate a dataset with preferences to run #DPO. This recent #NeurIPS paper answered some of my questions. The findings are simple but crucial for data creation:

πŸŽ“ Come and join our group! πŸŽ“
We offer 2 fully funded PhD positions:
🌍 Human-Centred Evaluation Frameworks for Multilingual Technologies (A6)
πŸ€– Multimedia Personalization with Multimodal Large Language Models (A7)
⏰ Deadline: 15 May 2026
πŸ”— Details: https://iecs.unitn.it/education/admission/call-for-application

Load More