Aakash Lahoti

I am currently a Ph.D. student in the Machine Learning Department at Carnegie Mellon University, advised by Prof. Albert Gu. I am interested in developing efficient and theoretically-grounded state space models (SSMs) along with their applications across domains.

Previously, I have worked on efficient Text-to-Speech models at Cartesia AI, developed content safety classifiers at Meta, and contributed to Google Assistant’s media backend at Google.

Please feel free to email me if you want to talk research and collaborate 😊

🎓 Education

  • Carnegie Mellon University, Pittsburgh, USA
    Master of Science (M.S.) in Machine Learning (2021 – 2022)
  • Indian Institute of Technology Kanpur (IIT Kanpur), India
    Bachelor of Technology (B.Tech.) in Computer Science and Engineering (2016 – 2020)

📢 News

  • [2025.02] 👀 Keep an eye out for Chimera that elegantly generalizes SSMs to graphs!
  • [2024.07] 😊 We are releasing Hydra’s • [Paper] • [Blog] • [Code]
  • [2024.07] 🎉 Our work Hydra has been accepted to NeurIPS 2024 • [Paper]
  • [2024.03] 🎉 Our work Role of Locality and Weight Sharing in Image-Based Tasks accepted to ICLR 2024 as a spotlight paper • [Paper]
  • [2024.03] 🎉 Our work Sharpened Lazy Incremental Quasi-Newton Method accepted to AISTATS 2024 • [Paper]