iiinova — where your digital self keeps evolving

Germán Kruszewski

@germank.bsky.social

Senior Scientist @Naver Labs Europe. MSCA Postdoctoral Research @UPF (COLT). Just trying out bsky for now...

Germán Kruszewski @germank.bsky.social · 18/03/2026

Inference-time scaling is great, but it only works if your model actually explores. RL training — by default — kills that exploration. If you are curious to learn why and what we did about it in about our recent ICLR 2026 paper, check out this 🧵!

Our trained models form a Pareto frontier between coverage and precision

141

Reposted by Germán Kruszewski

Thibaut Thonet @tthonet.bsky.social · 31/01/2025

🚨 Excited about ML/NLP and looking for a research internship on controlled text generation? Come work with us on advanced constraint processing in large language models at NAVER LABS Europe! ✨ Learn more and apply here: europe.naverlabs.com/job/2-7/

europe.naverlabs.com

Internship: Advanced Constraint Processing in LLMs

The ability to control the outputs of Large Language Models is a central topic in NLP and Machine Learning, with applications to safety, trustworthiness, reasoning, etc. Following […]

084