Queremos agradecer de nuevo el maravilloso patrocinio de cohere Labs al #HackathonSomosNLP 2025: $500 de su API para cada equipo 🔥 🔥 🔥 Hoy aprenderemos cómo utilizarla con Alejandro. ¡Únete a y pregúntale cualquier duda sobre la API y Cohere! ✅ forms.gle/bDaBC7XV3iu2tr…

It is critical for scientific integrity that we trust our measure of progress. The lmarena.ai has become the go-to evaluation for AI progress. Our release today demonstrates the difficulty in maintaining fair evaluations on lmarena.ai, despite best intentions.

Are there any tools that given a subword (BPE, SentencePiece, etc) tokenizer vocabulary, can infer what languages were likely used to train it? If not, sounds like a fun project! 👀

⚠️ Attention: The site is currently down. Our engineering team is investigating. We will update as soon as possible. You can track progress here: status.overleaf.com Sorry for any inconvenience.

I'm excited to share our new pre-print ShiQ: Bringing back Bellman to LLMs! arxiv.org/abs/2505.11081 In this work, we propose a new, Q-learning inspired RL algorithm for finetuning LLMs 🎉 (1/n)

Y yo que pensaba que no podía ser más fan de Carles Tamayo youtube.com/watch?v=xq3XXW…

🔠 UTF-8 was never meant for language models. Yet every major tokenizer still uses it, creating unfair "byte premiums". Why should your native script cost more to tokenize? It's time for a change. 🧵👇

How can we make language models more flexible to adapt to new languages after pretraining? 🌏 🧠 Our latest work investigates whether a tokenizer trained on more languages than the pretraining target can improve language plasticity without compromising pretraining performance.

🚨New pretraining paper on multilingual tokenizers 🚨 Super excited to share my work with Cohere Labs: One Tokenizer To Rule Them All: Emergent Language Plasticity via Multilingual Tokenizers

Oye Vodafone España hay gente llamando en vuestro nombre con datos de personas que no son clientes. Y cuando uno pregunta que le aclaren de qué compañía llaman, responden de malas formas diciendo que “lo han dejado claro y que no les haga perder tiempo”. ¿Esto dónde os lo denuncio?

Many colleagues from cohere and Cohere Labs will be in Vienna next week. Go talk to them and learn about our research. They're pretty nice people! You can even share a piece of Sacher cake together and discuss some papers! 🧠

Applications are now open for the next cohort of the Cohere Labs Scholars Program! 🌟 This is your chance to collaborate with some of the brightest minds in AI & chart new courses in ML research. Let's change the spaces breakthroughs happen. Apply by Aug 29.

¿Estás buscando una nueva oportunidad? ¿Te gustan los proyectos abiertos que desarrolla Cohere Labs? ¡Aplica a su Research Scholars Program antes del 29 de agosto!

We're searching for the next wave of rising stars ✨ Applications are open until the 29th: cohere.com/research/schol…

Ven a hacer investigación en IA con nosotros!

🚨1 day left to apply! Don’t miss your chance to jump in the deep end of ML research, with the resources and mentorship to succeed. Worldwide. Remote. Full-time. Paid. cohere.com/research/schol…