papers

Publications (25)

cs.CV2020

Thanks for Nothing: Predicting Zero-Valued Activations with Lightweight Convolutional Neural Networks

Gil Shomron, Ron Banner, Moran Shkolnik +1

Convolutional neural networks (CNNs) introduce state-of-the-art results for various tasks with the price of high computational demands. Inspired by the observation that spatial cor…

cs.CL2026

Nemotron 3 Ultra: Open, Efficient Mixture-of-Experts Hybrid Mamba-Transformer Model for Agentic Reasoning

NVIDIA, :, Aaron Blakeman +571

We introduce Nemotron 3 Ultra, a 550 billion total and 55 billion active parameter Mixture-of-Experts Hybrid Mamba-Attention language model. We pre-trained Nemotron 3 Ultra on 20 t…

cs.AI2021

Accelerated Sparse Neural Training: A Provable and Efficient Method to Find N:M Transposable Masks

Itay Hubara, Brian Chmiel, Moshe Island +3

Unstructured pruning reduces the memory footprint in deep neural networks (DNNs). Recently, researchers proposed different types of structural pruning intending to reduce also the…

cs.CV2019

CAT: Compression-Aware Training for bandwidth reduction

Chaim Baskin, Brian Chmiel, Evgenii Zheltonozhskii +3

Convolutional neural networks (CNNs) have become the dominant neural network architecture for solving visual processing tasks. One of the major obstacles hindering the ubiquitous u…

cs.LG2020

Robust Quantization: One Model to Rule Them All

Moran Shkolnik, Brian Chmiel, Ron Banner +4

Neural network quantization methods often involve simulating the quantization process during training, making the trained model highly dependent on the target bit-width and precise…

cs.LG2024

Accurate Neural Training with 4-bit Matrix Multiplications at Standard Formats

Brian Chmiel, Ron Banner, Elad Hoffer +2

Quantization of the weights and activations is one of the main methods to reduce the computational footprint of Deep Neural Networks (DNNs) training. Current methods enable 4-bit q…

cs.CV2019

Feature Map Transform Coding for Energy-Efficient CNN Inference

Brian Chmiel, Chaim Baskin, Ron Banner +5

Convolutional neural networks (CNNs) achieve state-of-the-art accuracy in a variety of tasks in computer vision and beyond. One of the major obstacles hindering the ubiquitous use…

cs.LG2020

Improving Post Training Neural Quantization: Layer-wise Calibration and Integer Programming

Itay Hubara, Yury Nahshan, Yair Hanani +2

Lately, post-training quantization methods have gained considerable attention, as they are simple to use, and require only a small unlabeled calibration set. This small dataset can…

cs.LG2024

Minimum Variance Unbiased N:M Sparsity for the Neural Gradients

Brian Chmiel, Itay Hubara, Ron Banner +1

In deep learning, fine-grained N:M sparsity reduces the data footprint and bandwidth of a General Matrix multiply (GEMM) up to x2, and doubles throughput by skipping computation of…

cs.LG2023

DropCompute: simple and more robust distributed synchronous training via compute variance reduction

Niv Giladi, Shahar Gottlieb, Moran Shkolnik +5

Background: Distributed training is essential for large scale training of deep neural networks (DNNs). The dominant methods for large scale DNN training are synchronous (e.g. All-R…

cs.LG2023

Graph Representation Learning via Aggregation Enhancement

Maxim Fishman, Chaim Baskin, Evgenii Zheltonozhskii +3

Graph neural networks (GNNs) have become a powerful tool for processing graph-structured data but still face challenges in effectively aggregating and propagating information betwe…

cs.AI2026

Workspace Optimization: How to Train Your Agent

Elad Sarafian, Gal Kaplun, Ron Banner +2

Modern agents built on frontier language models often cannot adapt their weights. What, then, remains trainable? We argue it is the agent's \emph{workspace}, the structured externa…

cs.LG2024

EXAQ: Exponent Aware Quantization For LLMs Acceleration

Moran Shkolnik, Maxim Fishman, Brian Chmiel +3

Quantization has established itself as the primary approach for decreasing the computational and storage expenses associated with Large Language Models (LLMs) inference. The majori…

cs.LG2022

Energy awareness in low precision neural networks

Nurit Spingarn Eliezer, Ron Banner, Elad Hoffer +2

Power consumption is a major obstacle in the deployment of deep neural networks (DNNs) on end devices. Existing approaches for reducing power consumption rely on quite general prin…

cs.LG2026

Normalized Architectures are Natively 4-Bit

Maxim Fishman, Brian Chmiel, Ron Banner +2

Training large language models at 4-bit precision is critical for efficiency. We show that nGPT, an architecture that constrains weights and hidden representations to the unit hype…

cs.CV2021

GAN "Steerability" without optimization

Nurit Spingarn-Eliezer, Ron Banner, Tomer Michaeli

Recent research has shown remarkable success in revealing "steering" directions in the latent spaces of pre-trained GANs. These directions correspond to semantically meaningful ima…

stat.ML2019

Norm matters: efficient and accurate normalization schemes in deep networks

Elad Hoffer, Ron Banner, Itay Golan +1

Over the past few years, Batch-Normalization has been commonly used in deep networks, allowing faster training and high performance for a wide variety of applications. However, the…

cs.LG2025

FP4 All the Way: Fully Quantized Training of LLMs

Brian Chmiel, Maxim Fishman, Ron Banner +1

We demonstrate, for the first time, fully quantized training (FQT) of large language models (LLMs) using predominantly 4-bit floating-point (FP4) precision for weights, activations…

cs.CV2020

Neural gradients are near-lognormal: improved quantized and sparse training

Brian Chmiel, Liad Ben-Uri, Moran Shkolnik +3

While training can mostly be accelerated by reducing the time needed to propagate neural gradients back throughout the model, most previous works focus on the quantization/pruning…

cs.LG2018

Scalable Methods for 8-bit Training of Neural Networks

Ron Banner, Itay Hubara, Elad Hoffer +1

Quantized Neural Networks (QNNs) are often used to improve network efficiency during the inference phase, i.e. after the network has been trained. Extensive research in the field s…

cs.LG2026

Retrieval from Within: An Intrinsic Capability of Attention-Based Models

Elad Hoffer, Yochai Blau, Edan Kinderman +3

Retrieval-augmented generation (RAG) typically treats retrieval and generation as separate systems. We ask whether an attention-based encoder-decoder can instead retrieve directly…

cs.DC2025

Distributed Training under Packet Loss

Erez Weintraub, Ron Banner, Ariel Orda

State-of-the-art language and vision models are routinely trained across thousands of GPUs, often spanning multiple data-centers, yet today's distributed frameworks still assume re…

cs.LG2025

Scaling FP8 training to trillion-token LLMs

Maxim Fishman, Brian Chmiel, Ron Banner +1

We train, for the first time, large language models using FP8 precision on datasets up to 2 trillion tokens -- a 20-fold increase over previous limits. Through these extended train…

cs.CV2019

Post-training 4-bit quantization of convolution networks for rapid-deployment

Ron Banner, Yury Nahshan, Elad Hoffer +1

Convolutional neural networks require significant memory bandwidth and storage for intermediate computations, apart from substantial computing resources. Neural network quantizatio…

cs.LG2020

Loss Aware Post-training Quantization

Yury Nahshan, Brian Chmiel, Chaim Baskin +4

Neural network quantization enables the deployment of large models on resource-constrained devices. Current post-training quantization methods fall short in terms of accuracy for I…