2025

1. 

   Accelerating Training with Neuron Interaction and Nowcasting Networks

   Boris Knyazev, Abhinav Moudgil, Guillaume Lajoie, Eugene Belilovsky, and 1 more author

   In ICLR, 2025

   arXiv Bib Code

   @inproceedings{knyazev2024accelerating,
     title = {Accelerating Training with Neuron Interaction and Nowcasting Networks},
     author = {Knyazev, Boris and Moudgil, Abhinav and Lajoie, Guillaume and Belilovsky, Eugene and Lacoste-Julien, Simon},
     booktitle = {ICLR},
     year = {2025},
   }

2. 

   Celo: Training Versatile Learned Optimizers on a Compute Diet

   Abhinav Moudgil, Boris Knyazev, Guillaume Lajoie, and Eugene Belilovsky

   Transactions on Machine Learning Research (TMLR), 2025

   arXiv Bib Code

   @article{moudgil2025celo,
     title = {Celo: Training Versatile Learned Optimizers on a Compute Diet},
     author = {Moudgil, Abhinav and Knyazev, Boris and Lajoie, Guillaume and Belilovsky, Eugene},
     journal = {Transactions on Machine Learning Research (TMLR)},
     year = {2025},
   }

3. 

   (Almost) Free Modality Stitching of Foundation Models

   Jaisidh Singh, Diganta Misra, Boris Knyazev, and Antonio Orvieto

   In EMNLP, 2025

   arXiv Bib

   @inproceedings{singh2025almost,
     title = {(Almost) Free Modality Stitching of Foundation Models},
     author = {Singh, Jaisidh and Misra, Diganta and Knyazev, Boris and Orvieto, Antonio},
     booktitle = {EMNLP},
     pages = {19784--19800},
     year = {2025},
   }

2024

1. 

   μLO: Compute-Efficient Meta-Generalization of Learned Optimizers

   Benjamin Thérien, Charles-Étienne Joseph, Boris Knyazev, Edouard Oyallon, and 2 more authors

   In OPT for ML 2024 NeurIPS Workshop, 2024

   arXiv Bib Code

   @inproceedings{therien2024mu,
     title = {$\mu${LO}: Compute-Efficient Meta-Generalization of Learned Optimizers},
     author = {Th{\'e}rien, Benjamin and Joseph, Charles-{\'E}tienne and Knyazev, Boris and Oyallon, Edouard and Rish, Irina and Belilovsky, Eugene},
     booktitle = {OPT for ML 2024 NeurIPS Workshop},
     year = {2024},
   }

2. 

   Graph Neural Networks for Learning Equivariant Representations of Neural Networks

   Miltiadis Kofinas, Boris Knyazev, Yan Zhang, Yunlu Chen, and 4 more authors

   In ICLR, 2024

   arXiv Bib Code

   @inproceedings{kofinas2024graph,
     title = {{G}raph {N}eural {N}etworks for {L}earning {E}quivariant {R}epresentations of {N}eural {N}etworks},
     author = {Kofinas, Miltiadis and Knyazev, Boris and Zhang, Yan and Chen, Yunlu and Burghouts, Gertjan J. and Gavves, Efstratios and Snoek, Cees G. M. and Zhang, David W.},
     booktitle = {ICLR},
     year = {2024},
   }

2023

1. 

   Can We Scale Transformers to Predict Parameters of Diverse Imagenet Models?

   Boris Knyazev, Doha Hwang, and Simon Lacoste-Julien

   In ICML, 2023

   arXiv Bib Code

   @inproceedings{knyazev2023can,
     title = {Can We Scale Transformers to Predict Parameters of Diverse Imagenet Models?},
     author = {Knyazev, Boris and Hwang, Doha and Lacoste-Julien, Simon},
     booktitle = {ICML},
     year = {2023},
   }

2022

1. 

   Hyper-representations as Generative Models: Sampling Unseen Neural Network Weights

   Konstantin Schürholt, Boris Knyazev, Xavier Giró-i-Nieto, and Damian Borth

   In NeurIPS, 2022

   arXiv Bib Code

   @inproceedings{schurholt2022hyper,
     title = {Hyper-representations as Generative Models: Sampling Unseen Neural Network Weights},
     author = {Sch{\"u}rholt, Konstantin and Knyazev, Boris and Gir{\'o}-i-Nieto, Xavier and Borth, Damian},
     booktitle = {NeurIPS},
     year = {2022},
   }

2. 

   On Evaluation Metrics for Graph Generative Models

   Rylee Thompson, Boris Knyazev, Elahe Ghalebi, Jungtaek Kim, and 1 more author

   In ICLR, 2022

   arXiv Bib Code

   @inproceedings{thompsone2022valuation,
     title = {On Evaluation Metrics for Graph Generative Models},
     author = {Thompson, Rylee and Knyazev, Boris and Ghalebi, Elahe and Kim, Jungtaek and Taylor, Graham W},
     booktitle = {ICLR},
     year = {2022},
   }

2021

1. 

   Parameter Prediction for Unseen Deep Architectures

   Boris Knyazev, Michal Drozdzal, Graham W Taylor, and Adriana Romero Soriano

   In NeurIPS, 2021

   arXiv Bib Code

   @inproceedings{knyazev2021parameter,
     title = {Parameter Prediction for Unseen Deep Architectures},
     author = {Knyazev, Boris and Drozdzal, Michal and Taylor, Graham W and Romero Soriano, Adriana},
     booktitle = {NeurIPS},
     year = {2021},
   }

2. 

   Brick-by-brick: Combinatorial Construction with Deep Reinforcement Learning

   Hyunsoo Chung, Jungtaek Kim, Boris Knyazev, Jinhwi Lee, and 3 more authors

   NeurIPS, 2021

   arXiv Bib Code

   @article{chung2021brick,
     title = {Brick-by-brick: Combinatorial Construction with Deep Reinforcement Learning},
     author = {Chung, Hyunsoo and Kim, Jungtaek and Knyazev, Boris and Lee, Jinhwi and Taylor, Graham W and Park, Jaesik and Cho, Minsu},
     journal = {NeurIPS},
     volume = {34},
     pages = {5745--5757},
     year = {2021},
   }

3. 

   Generative Compositional Augmentations for Scene Graph Prediction

   Boris Knyazev, Harm Vries, Cătălina Cangea, Graham W. Taylor, and 2 more authors

   In ICCV, 2021

   arXiv Bib Code

   @inproceedings{knyazev2020generative,
     title = {Generative Compositional Augmentations for Scene Graph Prediction},
     author = {Knyazev, Boris and de Vries, Harm and Cangea, Cătălina and Taylor, Graham W. and Courville, Aaron and Belilovsky, Eugene},
     booktitle = {ICCV},
     year = {2021},
   }

4. 

   Context-aware Scene Graph Generation with Seq2Seq Transformers

   Yichao Lu, Himanshu Rai, Jason Chang, Boris Knyazev, and 4 more authors

   In ICCV, 2021

   Bib HTML Code

   @inproceedings{lu2021context,
     title = {Context-aware Scene Graph Generation with Seq2Seq Transformers},
     author = {Lu, Yichao and Rai, Himanshu and Chang, Jason and Knyazev, Boris and Yu, Guangwei and Shekhar, Shashank and Taylor, Graham W and Volkovs, Maksims},
     booktitle = {ICCV},
     pages = {15931--15941},
     year = {2021},
   }

5. 

   Learning Temporal Attention in Dynamic Graphs with Bilinear Interactions

   Boris Knyazev, Carolyn Augusta, and Graham W Taylor

   Plos one, 2021

   arXiv Bib Code

   @article{knyazev2021learning,
     title = {Learning Temporal Attention in Dynamic Graphs with Bilinear Interactions},
     author = {Knyazev, Boris and Augusta, Carolyn and Taylor, Graham W},
     journal = {Plos one},
     volume = {16},
     number = {3},
     pages = {e0247936},
     year = {2021},
     publisher = {Public Library of Science San Francisco, CA USA},
   }

2020

1. 

   Graph Density-Aware Losses for Novel Compositions in Scene Graph Generation

   Boris Knyazev, Harm Vries, Cătălina Cangea, Graham W Taylor, and 2 more authors

   In British Machine Vision Conference (BMVC), 2020

   arXiv Bib Code

   @inproceedings{knyazev2020graphdensity,
     title = {Graph Density-Aware Losses for Novel Compositions in Scene Graph Generation},
     author = {Knyazev, Boris and de Vries, Harm and Cangea, Cătălina and Taylor, Graham W and Courville, Aaron and Belilovsky, Eugene},
     booktitle = {British Machine Vision Conference (BMVC)},
     year = {2020},
   }

2019

1. 

   Understanding Attention and Generalization in Graph Neural Networks

   Boris Knyazev, Graham W Taylor, and Mohamed Amer

   In NeurIPS, 2019

   arXiv Bib Code

   @inproceedings{knyazev2019understanding,
     title = {Understanding Attention and Generalization in Graph Neural Networks},
     author = {Knyazev, Boris and Taylor, Graham W and Amer, Mohamed},
     booktitle = {NeurIPS},
     year = {2019},
   }