Loading timeline…
20232020s
Organizations
Carnegie Mellon UniversityCartesia AI
Major Achievements
- •Developed State Space Models (SSMs) as a computationally cheaper alternative to transformers.
- •Co-authored Mamba (2023), achieving transformer-level performance but with linear-time scaling.
- •His work enabled models to easily ingest massive contexts like an entire hour of raw audio.
Key Papers
- Mamba: Linear-Time Sequence Modeling with Selective State Spaces
Albert Gu, Tri Dao · arXiv · 2023
- Efficiently Modeling Long Sequences with Structured State Spaces
Albert Gu, Karan Goel, Christopher Ré · ICLR 2022 · 2021