📢 We are excited to share that NLPnorth will present 4 papers at LREC COLING 2024 2024 in Turin! Find more about each of them in the 🧵 #NLProc

📢 NoDaLiDa/Baltic-HLT 2025 *call for papers* is up! 📢 Paper Deadline: *21 October 2024* Find more information on our ✨brand new website✨: nodalida-bhlt2025.eu/call-for-papers 165/166 days until deadline ░░░░░░░░░░░░░░░░░░░░ 0.6% #NLProc #nodalida_baltichlt

📝‼️Call for Workshop Proposals NoDaLiDa/Baltic-HLT 2025‼️📝 🗓️ 2 and 5 March, before and after the main conference. Options are full or half-day workshops. Deadline: 26 August 2024. Find more information on our website: nodalida-bhlt2025.eu/call-for-works…

The Aya Dataset now has a home and is accepted to #ACL2024. ✨ #NLProc

While the lovely folks at Cohere For AI will be mainly presenting the work, I will talk a bit about my time as a language ambassador for Dutch. Catch us at the poster session today at 16:00!! #NLProc #ACL2024NLP AAU TECH Department of Computer Science, Aalborg University

It was an absolute privilege to represent the 3,000 people around the 🗺️ that made Project Aya possible at Cohere For AI 💙 I’m very fortunate to have the opportunity to work with such fantastic people on problems that matter 🤗

📣 Two weeks until the paper submission deadline (21 October) 📣 We have reached the submission month and we're hoping to see all your work submitted 😄 Happy writing everyone ✍️!! More info here: nodalida-bhlt2025.eu/call-for-papers #NLProc

📣 Attention! 📣 ⚠️⚠️⚠️After careful consideration, the NoDaLiDa/Baltic-HLT Deadline is extended by one week. The new deadline is October 27 (Sunday), 23:59 anywhere on earth. ⚠️⚠️⚠️ #NLProc

Countdown begins 🔥✨ 🌎🌍🌏

Today I learned 👀🤫 #mysterybot

Frituur specialist! Probeer het zelf op **Whatsapp**: +1 (431) 302-8498 (Try it yourself with your own language on WhatsApp!) #mysterybot 🤫👀

As we prepare for Aya Expanse by Cohere For AI, I had a chance to try out the model in advance. From a spot check it works well for Dutch 👀. Test it out!! Dutch SMS: (+31) 97006520757 Whatsapp: +1 (431) 302-8498 #mysterybot

Interested in a PhD position in #NLProc at MaiNLPlab in beautiful Munich working on topics related to human-centric trustworthy NLP/AI/LLMs? Then hurry up, two weeks left to apply to our PIs: - #MCML mcml.ai/opportunities/… - ELLIS: This year, if you are applying to

🚀 Introducing INCLUDE 🌍: A multilingual LLM evaluation benchmark spanning 44 languages! Contains *newly-collected* data, prioritizing *regional knowledge*. Setting the stage for truly global AI evaluation. Ready to see how your model measures up? #AI #Multilingual #LLM #NLProc

🌟 Two years after ChatGPT's release, while the AI community focuses on advanced reasoning capabilities in English with models like o1, multilingual development deserves more attention! 🚀 We proudly present Sailor2, a community-driven project delivering state-of-the-art

Very happy to have learned about my old NLPnorth Mike Zhang 🦋 et al.'s work on SnakModel, a new language model for Danish!

🚀 Excited to share our technical report on the Southeast Asian multilingual model Sailor2 and its latest updates! Our 49-page report details Sailor2's development journey, including multilingual data cleaning, small model data mixture simulations, multi-stage continual

🚀 We are excited to introduce Kaleidoscope, the largest culturally-authentic exam benchmark. 📌 Most VLM benchmarks are English-centric or rely on translations—missing linguistic & cultural nuance. Kaleidoscope expands in-language multilingual 🌎 & multimodal 👀 VLMs evaluation

Very proud to introduce Kaleidoscope ✨🌿 🌍 18 languages (Bengali → Spanish) 📚 14 subjects (Humanities → STEM) 📸 55% requiring image understanding! A very important open science collaboration — which extends in-language evaluation for vision models to many more languages.

Thanks for sharing our work AK! We look at whether test-time compute improves other domains than math and science questions. In this case open-domain QA/factual QA. We show that up to a certain model size it gives substantial gains. /w Johannes Bjerva @ Russa Biswas