Zhao Song
- Papers
- 9
Cite
Notes
Only stored in your browser.
Authored papers
9LazyDiT: Lazy Learning for the Acceleration of Diffusion Transformers
arXiv 2024
Grams: Gradient Descent with Adaptive Momentum Scaling
arXiv 2024
Towards Infinite-Long Prefix in Transformer
arXiv 2024
Transformers are Deep Optimizers: Provable In-Context Learning for Deep Model Training
arXiv 2024
H$_2$O: Heavy-Hitter Oracle for Efficient Generative Inference of Large Language Models
arXiv 2023
Deja Vu: Contextual Sparsity for Efficient LLMs at Inference Time
arXiv 2023
Pixelated Butterfly: Simple and Efficient Sparse training for Neural Network Models
pixelated-butterfly-simple-and-efficient
Scatterbrain: Unifying Sparse and Low-rank Attention Approximation
scatterbrain-unifying-sparse-and-low-rank-1
WaveFlow: A Compact Flow-based Model for Raw Audio
ICML 2020 1
Affiliations
Frequent co-authors
10from 9 papers