Daniel Soudry
- Papers
- 5
Cite
Notes
Only stored in your browser.
5papers
Authored papers
5FP4 All the Way: Fully Quantized Training of LLMs
arXiv 2025
Scaling FP8 training to trillion-token LLMs
arXiv 2024
Alias-Free Convnets: Fractional Shift Invariance via Polynomial Activations
CVPR 2023 1
Improving Post Training Neural Quantization: Layer-wise Calibration and Integer Programming
arXiv 2020
Train longer, generalize better: closing the generalization gap in large batch training of neural networks
train-longer-generalize-better-closing-the-1
Affiliations
No known affiliations.
Frequent co-authors
9from 5 papers