Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models

Sanjiv Kumar is a Google Fellow and vice president at Google Research. His public homepage says he leads teams working on large machine learning foundation models and generative AI, has spent more than 25 years building machine learning systems and products, and received a PhD in computer science from Carnegie Mellon University in 2005.

Mona Jalal is a computer vision research engineer at Toyota Material Handling. Her homepage says she leads 3D computer vision work such as 6D object pose estimation and synthetic data generation, previously worked as an R&D engineer at UC Berkeley's FHL VIVE Center, completed doctoral study in computer vision at Boston University, and earlier earned master's degrees in computer sciences and electrical engineering from the University of Wisconsin-Madison.

Ming Lin is a professor of computer science at the University of Maryland, College Park and the chief executive officer and co-founder of Aitrios.ai. Her public homepage says she was the founding director of the GAMMA lab, previously served as assistant dean of the University of Maryland's Institute for Advanced Computer Studies, and earned her BS, MS, and PhD in EECS from the University of California, Berkeley.

Vice President of Applied Deep Learning Research at NVIDIA, leading work on conversational AI, generative AI, and accelerated deep learning software.

Senior research scientist at NVIDIA focused on large language models, reinforcement learning, and inference-time scaling for AI agents. His public NVIDIA author page also notes prior work on robotic grasping, pose estimation, and language understanding.

Ching-Yao Chuang is a researcher at OpenAI working on multimodal and generative AI systems. His homepage lists prior roles at xAI and Meta GenAI and a PhD from MIT.

Ming-Yu Liu is a vice president of research at NVIDIA and an IEEE Fellow. He leads the Deep Imagination Research group, which focuses on deep generative models for content creation and foundation models for physical AI. He previously held research leadership roles at Mercedes-Benz Research and Development North America and National Tsing Hua University, and he earned degrees from National Taiwan University, National Chiao Tung University, and the University of Illinois Urbana-Champaign.

Jinliang Guo is listed as an author of the NVIDIA technical report Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models.

Xiaosi Yang is listed as an author of the NVIDIA technical report Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models.

Junxian He is an assistant professor in computer science and engineering at the Hong Kong University of Science and Technology. His public homepage and Princeton profile say he works at the intersection of natural language processing and machine learning, focusing on scalable methods for language model training and reasoning, completed a PhD in computer science at Princeton University in 2024, and previously earned bachelor's degrees in economics and computer science from Peking University.

Rudra Murthy is listed as an author of the NVIDIA technical report Nemotron-H: A Family of Accurate and Efficient Hybrid Mamba-Transformer Models.

Arash Vahdat is a research director at NVIDIA Research, where he leads the GenAIR team on fundamental generative AI research. Before joining NVIDIA, he was a research scientist at D-Wave Systems and a research faculty member at Simon Fraser University, where he led video analysis research and taught machine learning for big data. He earned a PhD in computing science from Simon Fraser University.

Rogerio Feris is a principal scientist and manager at the MIT-IBM Watson AI Lab within IBM Research. His public bios emphasize multimodal perception, data and model efficiency, and large language models with memory.

Canonical link

Sanjiv Kumar

Mona Jalal

Ming Lin

Bryan Catanzaro

Amin Firooz

Ching-Yao Chuang

Ming-Yu Liu

Jinliang Guo

Xiaosi Yang

Junxian He

Rudra Murthy

Arash Vahdat

Rogerio Feris