My homepage!
I am a Research Scientist at Google Research with a PhD in Statistical Machine Learning and over 14 years of experience in research and industry. My expertise spans various AI/ML domains, including Large Language Models (LLMs), Deep Generative Models, Computer Vision, Efficient AI, and Graph Neural Networks. My work has been published in top AI conferences in collaboration with leading researchers from Google Brain and DeepMind. Additionally, I have valuable industry experience in deep learning, NLP, and Foundation Models, with proficiency in Python, particularly PyTorch, JAX, and TensorFlow. I bring a unique combination of problem-solving, creativity, out-of-the-box thinking, a passion for learning, project management, and strong communication skills.
M. Karami, V. Mirrokni, “Lattice: Learning to Efficiently Compress the Memory” Under Review arXiv Preprint 2025
M. Karami, A. Ghodsi, “Auto-Regressive Masked Diffusion Model” Under Review 2025
Behrouz, A., Parviz, A., Karami, M., Sanford, C., Perozzi, B. and Mirrokni, V., “Best of Both Worlds: Advantages of Hybrid Graph Sequence Models.” 42nd Int. Conf. on Machine Learning (ICML 2025) paper
M. Karami, A. Behrouz, P.Kacham, V. Mirrokni, “Trellis: Learning to Compress Key-Value Memory in Attention Models.” Conference on Language Modeling (COLM) 2025
M. Karami, A. Behrouz, P.Zhong, R. Pascanu, V. Mirrokni, “Enhancing Sequence Modeling with Multi-Resolution State Space Models.” Conference on Language Modeling (COLM) 2025
Pătrăucean, V., He, X.O., Heyward, J., Zhang, C., Sajjadi, M.S., Muraru, G.C., Zholus, A., Karami, M., Goroshin, R., Chen, Y. and Osindero, S., “TRecViT: A Recurrent Video Transformer.” arXiv preprint arXiv:2412.14294. 2025 paper
M. Karami, “HiGen: Hierarchical Graph Generative Networks”, International Conference on Learning Representations (ICLR), (2024). paper code
M. Karami, I. Krawczuk, V. Cevher, “Multi-Resolution Graph Diffusion”, ICLR 2024 Workshop on Machine Learning for Genomics Explorations, (2024).