Get started
Author

Brian Chmiel

Loss aware post-training quantization
2021 · Yury Nahshan, Brian Chmiel, et al. · Machine Learning · 187
PDF
Robust Quantization: One Model to Rule Them All
2020 · Shkolnik, Moran, Chmiel, Brian, et al. · arXiv (Cornell University) · 57
PDF
Scaling FP8 training to trillion-token LLMs
2024 · Fishman, Maxim, Chmiel, Brian, et al. · arXiv (Cornell University) · 56
PDF
Accelerated Sparse Neural Training: A Provable and Efficient Method to Find N:M Transposable Masks
2021 · Hubara, Itay, Chmiel, Brian, et al. · arXiv (Cornell University) · 46
PDF
FP4 All the Way: Fully Quantized Training of LLMs
2025 · Chmiel, Brian, Fishman, Maxim, et al. · arXiv (Cornell University) · 43
PDF
Neural gradients are near-lognormal: improved quantized and sparse training
2020 · Chmiel, Brian, Ben-Uri, Liad, et al. · arXiv (Cornell University) · 39
PDF
Accurate Neural Training with 4-bit Matrix Multiplications at Standard Formats
2021 · Chmiel, Brian, Banner, Ron, et al. · arXiv (Cornell University) · 29
PDF
Feature Map Transform Coding for Energy-Efficient CNN Inference
2020 · Brian Chmiel, Chaim Baskin, et al. · International Joint Conference on Neural Networks (IJCNN) · 26
PDF
CAT: Compression-Aware Training for bandwidth reduction
2019 · Baskin, Chaim, Chmiel, Brian, et al. · arXiv (Cornell University) · 11
PDF
Minimum Variance Unbiased N:M Sparsity for the Neural Gradients
2022 · Chmiel, Brian, Hubara, Itay, et al. · arXiv (Cornell University) · 9
PDF
Adversarial robustness via noise injection in smoothed models
2022 · Yaniv Nemcovsky, Evgenii Zheltonozhskii, et al. · Applied Intelligence · 6
EXAQ: Exponent Aware Quantization For LLMs Acceleration
2024 · Shkolnik, Moran, Fishman, Maxim, et al. · arXiv (Cornell University) · 5
PDF
Bimodal-Distributed Binarized Neural Networks
2022 · Tal Rozen, Moshe Kimhi, et al. · Mathematics · 5
PDF
Colored Noise Injection for Training Adversarially Robust Neural Networks
2020 · Zheltonozhskii, Evgenii, Baskin, Chaim, et al. · arXiv (Cornell University) · 5
PDF
Towards Learning of Filter-Level Heterogeneous Compression of Convolutional Neural Networks
2019 · Zur, Yochai, Baskin, Chaim, et al. · arXiv (Cornell University) · 4
PDF
FP4 All the Way: Fully Quantized Training of Large Language Models
2025 · Brian Chmiel, Maxim Fishman, et al. · neural information processing systems · 3
Smoothed Inference for Adversarially-Trained Models
2019 · Nemcovsky, Yaniv, Zheltonozhskii, Evgenii, et al. · arXiv (Cornell University) · 3
PDF
Normalized Architectures are Natively 4-Bit
2026 · Fishman, Maxim, Chmiel, Brian, et al. · arXiv (Cornell University) · 2
PDF
Neural gradients are lognormally distributed: understanding sparse and quantized training.
2020 · Brian Chmiel, Liad Ben-Uri, et al. · arXiv (Cornell University) · 2
Dory1.0-4B: A Looped Normalized-GPT
2026 · Fishman, Maxim, Sun, Simeng, et al. · arXiv (Cornell University)
PDF
UNREAL: Unifying Retrieval and Long-Context with a Single Model
2026 · Kinderman, Edan, Hoffer, Elad, et al. · arXiv (Cornell University)
PDF
Works count: 21
Citation count: 538
H-index: 9
I10-index: 9
Year
Topic
Type