Hallelujah! I’m excited to share that I’ve been selected as a 2025 AI2050 Early Career Fellow by @Schmidtsciences This year’s fellows represent 42 institutions across eight countries, working to ensure AI benefits humankind. Learn more at:

1/ 🌍 How does mixing data from hundreds of languages affect LLM training? In our new paper "Revisiting Multilingual Data Mixtures in Language Model Pretraining" we revisit core assumptions about multilinguality using 1.1B-3B models trained on up to 400 languages. 🧵👇

🏆 We are incredibly honored to announce that our paper, "Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free" has received the NeurIPS 2025 Best Paper Award! A huge congratulations to our dedicated research team for pushing the boundaries

New Anthropic research: Natural emergent misalignment from reward hacking in production RL. “Reward hacking” is where models learn to cheat on tasks they’re given during training. Our new study finds that the consequences of reward hacking, if unmitigated, can be very serious.

For more on our results, read our blog post: https://t.co/GLV9GcgvO6 And read our paper:

I’m pleased to share the Second Key Update to the International AI Safety Report, which outlines how AI developers, researchers, and policymakers are approaching technical risk management for general-purpose AI systems. (1/5)

Our #NeurIPS2025 paper shows that even comparable monolingual tokenizers have different compression rates across languages. But by getting rid of whitespace tokenization and using a custom vocab size for each language, we can reduce token premiums. Preprint out now!

Come and join us to work on robust, efficient and trustworthy AI across languages and domains. You will be part of the ELLIS Institute Finland as one of the prestigious PIs in their top-level research unit. Application deadline is January 12:

Today, we are releasing the best open-weight LLM by a US company: Cogito v2.1 671B. On most industry benchmarks and our internal evals, the model performs competitively with frontier closed and open models, while being ahead of any US open model (such as the best versions of

The secret behind Gemini 3? Simple: Improving pre-training & post-training 🤯 Pre-training: Contra the popular belief that scaling is over—which we discussed in our NeurIPS '25 talk with @ilyasut and @quocleix—the team delivered a drastic jump. The delta between 2.5 and 3.0 is

📢 I'm recruiting PhD students at MPI!! Topics include: 1⃣ LLM factuality, reliable info synthesis and reasoning, personalization + applications in real-world inc. education, science 2⃣ Data-centric interpretability 3⃣Creativity in AI, esp scientific applications 🧵1/2

🚀 Job market post I’m a final-year PhD student at @kaistpr CS, looking for #postdoc/industry research position starting Fall 2026! I develop multilingual, multicultural language models and data, inspired by how humans use and learn language. 🔗 https://t.co/LkmT7inSbH

ICLR authors, want to check if your reviews are likely AI generated? ICLR reviewers, want to check if your paper is likely AI generated? Here are AI detection results for every ICLR paper and review from @pangramlabs! It seems that ~21% of reviews may be AI?

📢 AfricaNLP Second Call for Papers 📢 The 7th AfricaNLP Workshop at EACL 2026 (Rabat, Morocco) invites papers on Multilingual Multimodal LLMs for African languages. 👉 Submission deadline: December 19 2025 (AoE) More details: https://t.co/aBDcSRcJxJ #AfricaNLP #EACL2026 #NLP

full leaderboard here

AfroBench-Lite updated results for several frontier models, we cover 14 African languages and 7 tasks in the Lite version. #AfricaNLP @MasakhaneNLP Thanks to @OpenAI , @AnthropicAI and @googlecloud for the API credits. Courtesy @JessicaOjo__

Introducing Meta Omnilingual Automatic Speech Recognition (ASR), a suite of models providing ASR capabilities for over 1,600 languages, including 500 low-coverage languages never before served by any ASR system. While most ASR systems focus on a limited set of languages that are

Congratulations to the winners 🏆 Let's continue to make efforts to scale multilingual evaluation to many languages

Congratulations 🎉

The second is 7PERFECTION, a dataset for seven Nigerian languages!

Finally, we recognize the Hawaiian submission to the shared task. Thank you for your contributions!