publications

2025

1. EMNLP-W

   Synthetic Proofs with Tool-Integrated Reasoning: Contrastive Alignment for LLM Mathematics with Lean

   Mark Obozov, Michael Diskin, Aleksandr Beznosikov, and 2 more authors

   In Proceedings of The 3rd Workshop on Mathematical Natural Language Processing (MathNLP 2025), 2025

   DOI Bib HTML PDF

   

   @inproceedings{obozov2025synthetic,
     title = {Synthetic Proofs with Tool-Integrated Reasoning: Contrastive Alignment for LLM Mathematics with Lean},
     author = {Obozov, Mark and Diskin, Michael and Beznosikov, Aleksandr and Gasnikov, Alexander and Barannikov, Serguei},
     booktitle = {Proceedings of The 3rd Workshop on Mathematical Natural Language Processing (MathNLP 2025)},
     year = {2025},
     pages = {195--202},
     url = {https://aclanthology.org/2025.mathnlp-main.15/},
     doi = {10.18653/v1/2025.mathnlp-main.15},
   }

2. NeurIPS-W

   Think, Align, Select: Query–Key Scores for LLM Reasoning

   Mark Obozov, Eduard Tulchinskii, Kristian Kuznetsov, and 2 more authors

   In The 5th Workshop on Mathematical Reasoning and AI (NeurIPS), 2025

   Bib HTML PDF

   

   @inproceedings{obozov2025think,
     title = {Think, Align, Select: Query--Key Scores for LLM Reasoning},
     author = {Obozov, Mark and Tulchinskii, Eduard and Kuznetsov, Kristian and Diskin, Michael and Barannikov, Serguei},
     booktitle = {The 5th Workshop on Mathematical Reasoning and AI (NeurIPS)},
     year = {2025},
     url = {https://openreview.net/forum?id=VLY6qAX4Fe},
   }

2023

1. ICML

   SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient

   Max Ryabinin, Tim Dettmers, Michael Diskin, and 1 more author

   In International Conference on Machine Learning (ICML), 2023

   arXiv Bib HTML PDF

   

   @inproceedings{ryabinin2023swarm,
     title = {SWARM Parallelism: Training Large Models Can Be Surprisingly Communication-Efficient},
     author = {Ryabinin, Max and Dettmers, Tim and Diskin, Michael and Borzunov, Alexander},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2023},
     url = {https://arxiv.org/abs/2301.11913},
   }

2. ICLR

   A Critical Look at the Evaluation of GNNs under Heterophily: Are We Really Making Progress?

   Oleg Platonov, Denis Kuznedelev, Michael Diskin, and 2 more authors

   In International Conference on Learning Representations (ICLR), 2023

   arXiv Bib HTML PDF Code

   

   @inproceedings{platonov2023critical,
     title = {A Critical Look at the Evaluation of GNNs under Heterophily: Are We Really Making Progress?},
     author = {Platonov, Oleg and Kuznedelev, Denis and Diskin, Michael and Babenko, Artem and Prokhorenkova, Liudmila},
     booktitle = {International Conference on Learning Representations (ICLR)},
     year = {2023},
     url = {https://arxiv.org/abs/2302.11640},
   }

2022

1. NeurIPS

   Distributed Methods with Compressed Communication for Solving Variational Inequalities, with Theoretical Guarantees

   Aleksandr Beznosikov, Peter Richtárik, Michael Diskin, and 2 more authors

   In Advances in Neural Information Processing Systems (NeurIPS), 2022

   arXiv Bib HTML PDF

   

   @inproceedings{beznosikov2022distributed,
     title = {Distributed Methods with Compressed Communication for Solving Variational Inequalities, with Theoretical Guarantees},
     author = {Beznosikov, Aleksandr and Richt{\'a}rik, Peter and Diskin, Michael and Ryabinin, Max and Gasnikov, Alexander},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
     year = {2022},
     url = {https://arxiv.org/abs/2110.03313},
   }

2. ICML

   Secure Distributed Training at Scale

   Eduard Gorbunov, Alexander Borzunov, Michael Diskin, and 1 more author

   In International Conference on Machine Learning (ICML), 2022

   arXiv Bib HTML PDF

   

   @inproceedings{gorbunov2022secure,
     title = {Secure Distributed Training at Scale},
     author = {Gorbunov, Eduard and Borzunov, Alexander and Diskin, Michael and Ryabinin, Max},
     booktitle = {International Conference on Machine Learning (ICML)},
     year = {2022},
     url = {https://arxiv.org/abs/2106.11257},
   }

3. NeurIPS

   Training Transformers Together

   Alexander Borzunov, Max Ryabinin, Tim Dettmers, and 5 more authors

   In NeurIPS 2021 Competitions and Demonstrations Track, 2022

   arXiv Bib HTML PDF

   

   @inproceedings{borzunov2022training,
     title = {Training Transformers Together},
     author = {Borzunov, Alexander and Ryabinin, Max and Dettmers, Tim and Lhoest, Quentin and Saulnier, Lucile and Diskin, Michael and Jernite, Yacine and Wolf, Thomas},
     booktitle = {NeurIPS 2021 Competitions and Demonstrations Track},
     pages = {335--342},
     year = {2022},
     url = {https://arxiv.org/abs/2207.03481},
   }

2021

1. NeurIPS

   Distributed Deep Learning in Open Collaborations

   Michael Diskin, Alexey Bukhtiyarov, Max Ryabinin, and 13 more authors

   In Advances in Neural Information Processing Systems (NeurIPS), 2021

   arXiv Bib HTML PDF Code

   

   @inproceedings{diskin2021distributed,
     title = {Distributed Deep Learning in Open Collaborations},
     author = {Diskin, Michael and Bukhtiyarov, Alexey and Ryabinin, Max and Saulnier, Lucile and Lhoest, Quentin and Sinitsin, Anton and Popov, Dmitry and Pyrkin, Dmitry and Kashirin, Maxim and Borzunov, Alexander and Villanova del Moral, Albert and Mazur, Denis and Kobelev, Ilia and Jernite, Yacine and Wolf, Thomas and Pekhimenko, Gennady},
     booktitle = {Advances in Neural Information Processing Systems (NeurIPS)},
     year = {2021},
     url = {https://arxiv.org/abs/2106.10207},
   }

2020

1. OSS

   Hivemind: Decentralized Deep Learning in PyTorch

   Max Ryabinin, Alexander Borzunov, Michael Diskin, and 1 more author

   2020

   Bib HTML Code

   

   @misc{ryabinin2020hivemind,
     title = {Hivemind: Decentralized Deep Learning in PyTorch},
     author = {Ryabinin, Max and Borzunov, Alexander and Diskin, Michael and others},
     year = {2020},
     url = {https://github.com/learning-at-home/hivemind},
   }