About me

I am a natural language processing researcher at Naver Labs Europe in Grenoble, France, focusing on making recent NLP advances applicable across domains and languages. Together with colleagues, I am co-teaching an NLP course at CentraleSupelec and co-organizing an ALPS winter school.

Before joining NLE, I did my PhD on deep learning for source code processing at the Centre of Deep Learning and Bayesian Methods of the HSE University, where I was advised by Prof. Dmitry Vetrov, the head of our Bayesian Methods Research Group. I received my Bachelor's and Master's degrees at Moscow State University. At HSE, I also taught machine learning to Bachelor and Master students for 6 years and was selected by students for the Best Teacher award 3 times. During 2017-2019, I've been one of the main organizers of the Summer School on Deep Learning and Bayesian Methods.

News

• 2026
  • Publicity and Social Media chair @ EACL 2026
  • Presented LLM-as-a-qualitative-judge at a Multilingual Multicultural evaluation workshop @ EACL 2026 in Rabat, Morocco
  • Multilingual Provence, a work led by our intern Youssef Mohamed, was accepted to ECIR 2026
  • We organized the 6th Advanced Language Processing School taking place in the French Alps
• 2025
  • Area chair @ ACL rolling review
  • Presented Provence (Pruning and Reranking Of retrieVEd relevaNt ContExts) at ICLR 2025 in Singapore [blogpost] [open-source reproduction by the community] [Perplexity builds on top of Provence]
  • We organized the 5th Advanced Language Processing School taking place in the French Alps
• 2024
  • Presented our study on cross-lingual transfer in generative tasks at NAACL 2024 in Mexico City, and its follow-up on cross-lingual transfer in instruction tuning of LLMs at INLG 2024 in Tokyo, Japan
  • Our work on Bergen, a benchmarking library for RAG, has been accepted to EMNLP 2024 Findings. I also presented its multilingual extension and our study on multilingual RAG at the Knowledgeable LLMs workshop @ ACL'24 in Bangkok, Thailand
• 2023
  • Presented our work on ambiguity in tokenization at ACL 2023 in Toronto, Canada
  • Presented our CodeBPE paper at ICLR 2023 in Kigali, Rwanda
• 2022
  • Defended my PhD in Computer Science! (link to the thesis)
  • Joined Naver Labs Europe in Grenoble, France as an NLP researcher!
  • Presented our spotlight paper on CodeBPE at the DL4Code workshop of ICLR 2022 (online)
  • Gave a talk about our results in DL4Code at JetBrains research seminar (online)
• 2021
  • We presented our work on analyzing the interplay between weight decay and batch normalization at NeurIPS 2021 (online, with Ekaterina Lobacheva and Maxim Kodryan)
  • Presented our work Empirical Study of Transformers for Source Code at ESEC/FSE 2021 (online, with Sergey Troshin) [code]
  • Presented two papers on deep learning for source code at NAACL 2021 (online, a single-author long one and a short one with Sergey Troshin)
• 2020
  • We presented our spotlight paper On Power Laws in Deep Ensembles at NeurIPS 2020 (online, with Ekaterina Lobacheva and Maxim Kodryan)
  • Ekaterina Lobacheva and I presented our paper at AAAI Conference on Artificial Intelligence in New York, USA
• 2019
  • Presented our paper at NeurIPS Workshop on Context and Compositionality in Biological and Artificial Neural Systems in Vancouver, Canada
  • Gave a lecture on Bayesian neural networks at Machine Learning and Applications to Physics workshop in Madrid, Spain
  • Gave lecture and seminar on Bayesian neural networks at the Fifth Machine Learning in High Energy Physics Summer School in Hamburg, Germany
  • Gave a tutorial on Bayesian sparsification of neural networks at Probabilistic AI summer school in Trondheim, Norway [video]
  • Took part in Google NLP summit in Zurich, Switzerland
  • We organized the third Deep|Bayes summer school in Moscow, Russia.
• 2018
  • Gave a contributed talk presenting our paper at NeurIPS Workshop on Compact Deep Neural Networks with industrial applications, Montreal, Canada
  • Presented our paper at EMNLP 2018, Brussels, Belgium (with Ekaterina Lobacheva)
  • We organized the second Summer school on deep learning and Bayesian methods in Moscow, Russia. I was one of the main organizers
• 2017
  • Our research group organized the first Summer school on deep learning and Bayesian methods in Moscow, Russia. We with Kate Lobacheva ran all the organization
  • Pesented our paper at the ICML Workshop on Learning to Generate Natural Language, Sydney, Australia (with Ekaterina Lobacheva)
• 2016
  • Our paper was published in the Journal on machine learning and data analysis. I gave a talk presenting it at the 11th International Conference on Intelligent Data Processing in Barcelona, Spain