Staff Research Scientist at Cohere Labs. Research on multilingual and multicultural LLMs, reinforcement learning from human feedback, and evaluation.
I’m a Staff Research Scientist at Cohere Labs, where I conduct research on large language models, centered around multilinguality, reinforcement learning, and evaluation. I am also an associate industry member at MILA. I am based in Canada, in the suburbs of Montreal. Previously I worked at Google Translate (Research), Montreal, with a focus on machine translation. Very broadly speaking, I am interested in the intersection of natural language processing (NLP) and machine learning, where multiple languages come into play. I obtained my PhD from Heidelberg University, Germany, under supervision of Prof. Stefan Riezler at the StatNLP group.
In my PhD, I investigated how reinforcement learning algorithms can be used to turn weak supervision signals from users into meaningful updates for a machine translation system. If you’re into RLHF, check out my early works on learning from simulated human feedback (ACL 2016) - we called it “bandit structured prediction” back then, and learning from actual user feedback (NAACL 2018) for sequence-to-sequence models. You can find many elements (and challenges) of today’s leading RL algorithms there, just at smaller scale and before LLMs and the RLHF branding. A few years later we went back to these basics by taking apart PPO and bringing back good old REINFORCE. I gave a lecture at AthNLP 2026 to summarize these 10 years of RLHF (slides available on the AthNLP page).
🎯 My long-term goal for NLP research is to make it more accessible and inclusive, along multiple dimensions:
Check out our lab’s scholar program and Grant calls, these are unique opportunities for getting started with LLM research.
Most leading LLMs today are multilingual, more or less intentionally. I don’t see it as optional to evaluate techniques and models across languages, but rather as an obligation for responsible research with a global perspective (most critical for safety). Besides adding one dimension to every evaluation, it is a good reality check for robustness of developed techniques. While multilinguality can seen overwhelming and daunting, it is the perfect opportunity for collaboration, as we can put our global network of researchers into action, e.g. in data audits, or for open LLM or data development. More recently, I have focused on the aspect of culture that is found in the layer below language. Even if LLMs get the surface language right, there are large discrepancies for how well they are able to understand, relate to and generate suitable content for cultures around the globe.
The faster model development and releases, the more important become evaluations. They represent our compass and proxies to real-world impact. It is incredibly important that we make sure to constantly evolve them, and maintain rigor and the will to look beyond a single score or ranking. I have written about this in two blogposts (Elo ratings, fair and comprehensive multilingual LLM evaluation practices) in collaboration with AI Singapore, and in a recent COLM paper, where we draw the connections between LLM and MT evaluations.
⏳ Last updated: 20 Oct 2026. If there’s no recent news below, it means I have been too busy with life and research.
Email: <lowercase first + last name>@cohere.com
LinkedIn: linkedin.com/in/juliakreutzer
Please don’t let your agent email / message me. I’m allergic to AI slop and agent-authored email hooks, you don’t impress me by automation. I prefer authentic human-to-human communication with all its imperfections.