Computational Linguistics @ UZH

cl-uzh.bsky.social

10 April 2026 08:08

🔵 Apertus: Democratizing Open and Compliant LLMs for Global Language Environments. arxiv.org/abs/2509.14233 6/7

Computational Linguistics @ UZH

cl-uzh.bsky.social

10 April 2026 08:08

🔵 @negarforoutan.bsky.social, Clara Meister, Debjit Paul, @joelniklaus.bsky.social, @sinaahmadi.bsky.social, Antoine Bosselut, @ricosennrich.bsky.social . Parity-Aware Byte-Pair Encoding: Improving Cross-lingual Fairness in Tokenization. arxiv.org/abs/2508.04796 5/7

Leshem (Legend) Choshen @EMNLP

lchoshen.bsky.social

14 January 2026 19:56

"It is time to separate language from language models" The revelation keeps bugging me, so it also underlies the talk I just gave "multilingual?" Thought I'd briefly share the contents of the talk: 🤖📈🧠

Francesca Padovani

frap98.bsky.social

09 November 2025 07:38

𝐃𝐨 𝐲𝐨𝐮 𝐫𝐞𝐚𝐥𝐥𝐲 𝐰𝐚𝐧𝐭 𝐭𝐨 𝐬𝐞𝐞 𝐰𝐡𝐚𝐭 𝐦𝐮𝐥𝐭𝐢𝐥𝐢𝐧𝐠𝐮𝐚𝐥 𝐞𝐟𝐟𝐨𝐫𝐭 𝐥𝐨𝐨𝐤𝐬 𝐥𝐢𝐤𝐞? 🇨🇳🇮🇩🇸🇪 Here’s the proof! 𝐁𝐚𝐛𝐲𝐁𝐚𝐛𝐞𝐥𝐋𝐌 is the first Multilingual Benchmark of Developmentally Plausible Training Data available for 45 languages to the NLP community 🎉 arxiv.org/abs/2510.10159

Jaap Jumelet

jumelet.bsky.social

09 November 2025 07:36

🌍Introducing BabyBabelLM: A Multilingual Benchmark of Developmentally Plausible Training Data! LLMs learn from vastly more data than humans ever experience. BabyLM challenges this paradigm by focusing on developmentally plausible data We extend this effort to 45 new languages!