With @andrei-a-manea.bsky.social, we posted a survey on multilingual vision-language models 👉 arxiv.org/pdf/2509.22123 We reviewed 31 models+21 benchmarks. There's a tension between language neutrality (same results across languages) & cultural awareness (context matters differently across cultures)

Investigating the Effect of Parallel Data in the Cross-Lingual Transfer for Vision-Language Encoders by @andrei-a-manea.bsky.social & @jlibovicky.bsky.social TL;DR: Explore how parallel datasets improve cross-lingual transfer in vision-language models. arxiv.org/abs/2504.21681

Great week at #TSD2025 in Erlangen! Our @ufal-cuni.bsky.social team presented 7 papers covering various topics from Czech speech assessment to multilingual morphology. Thanks to all attendees who engaged with our work! 🧵👇 #NLP #ComputationalLinguistics #CzechNLP #MachineLearning

Throwback to ACL, another poster presentation is coming at TSD tomorrow. If you are around, let's meet there

Institute of Formal and Applied Linguistics@ufal.mff.cuni.cz · last yr.

CUS-QA: Local-Knowledge-Oriented Open-Ended Question Answering Dataset arxiv.org/abs/2507.22752 by @jlibovicky.bsky.social , ‪@jindrahelcl.bsky.social, @andrei-a-manea.bsky.social Question that foreigners don't know the answer to + human judgment on question generation

Our latest contribution to multilingual NLP evaluation 🚀 by @jlibovicky.bsky.social, @jindrahelcl.bsky.social, @andrei-a-manea.bsky.social & @gianlucavico.bsky.social

Jindřich Libovický@jlibovicky.bsky.social · 12mo ago

🧵 We're releasing CUS-QA - a new benchmark for testing LLMs on regional knowledge! Find out what your model knows about Czechia 🇨🇿, Slovakia 🇸🇰, and Ukraine 🇺🇦! 👉 Textual and visual questions, answers, and human judgment on model outputs! huggingface.co/datasets/ufa... www.arxiv.org/abs/2507.22752

#ACL2025NLP continues with two days of workshop, and @ufal-cuni.bsky.social folks are there with more than a dozen posters 👇