Scholar

Songlin Yang

Google Scholar ID: 1chlis0AAAAJ

MIT

machine learning

Citations & Impact

All-time

Citations

1,021

H-index

i10-index

Publications

Co-authors

list available

Contact

Publications

1 items

arXiv.org · 2026

Cited

Resume (English only)

Academic Achievements

First author of 'Gated Linear Attention Transformers with Hardware-Efficient Training' (ICML 2024)
First author of 'Gated Delta Networks: Improving Mamba2 with Delta Rule' (ICLR 2025)
First author of 'Parallelizing Linear Transformers with the Delta Rule over Sequence Length' (NeurIPS 2024)
First author of 'PaTH Attention: Position Encoding via Accumulating Householder Transformations' (NeurIPS 2025)
Developed and maintains the Flash Linear Attention open-source project and community

Co-authors

12 total