English
Related papers

Related papers: Adaptive Block-Scaled Data Types

200 papers

As large language models have grown larger, interest has grown in low-precision numerical formats such as NVFP4 as a way to improve speed and reduce memory usage. However, quantizing models to NVFP4 remains challenging as the lack of…

Computation and Language · Computer Science 2026-05-12 Jack Cook , Junxian Guo , Guangxuan Xiao , Yujun Lin , Keith Wyss , Mahdi Nazemi , Asit Mishra , Carlo del Mundo , Tijmen Blankevoort , Song Han

Quantization addresses the high resource demand for large language models (LLMs) by alleviating memory pressure and bandwidth congestion and providing significantly scaled compute power with a tolerable impact on accuracy. Four-bit floating…

Hardware Architecture · Computer Science 2026-03-11 Musa Cim , Burak Topcu , Mahmut Taylan Kandemir

This paper introduces HiFloat4 (HiF4), a block floating-point data format tailored for deep learning. Each HiF4 unit packs 64 4-bit elements with 32 bits of shared scaling metadata, averaging 4.5 bits per value. The metadata specifies a…

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, and gradients on datasets up to 200 billion tokens. We…

Machine Learning · Computer Science 2025-08-12 Brian Chmiel , Maxim Fishman , Ron Banner , Daniel Soudry

The rapidly increasing size of large language models (LLMs) presents significant challenges in memory usage and computational costs. Quantizing both weights and activations can address these issues, with hardware-supported fine-grained…

Computation and Language · Computer Science 2025-07-25 Wonsuk Jang , Thierry Tambe

Large language models (LLMs) demand extensive memory capacity during both fine-tuning and inference. To enable memory-efficient fine-tuning, existing methods apply block-wise quantization techniques, such as NF4 and AF4, to the network…

Machine Learning · Computer Science 2025-05-13 Patrick Blumenberg , Thomas Graave , Tim Fingscheidt

Conventional hardware-friendly quantization methods, such as fixed-point or integer, tend to perform poorly at very low word sizes as their shrinking dynamic ranges cannot adequately capture the wide data distributions commonly seen in…

Machine Learning · Computer Science 2020-02-12 Thierry Tambe , En-Yu Yang , Zishen Wan , Yuntian Deng , Vijay Janapa Reddi , Alexander Rush , David Brooks , Gu-Yeon Wei

Deploying large language models (LLMs) on edge devices requires extremely low-bit quantization. Ultra-low precision formats such as NVFP4 offer a promising solution for reducing memory footprint and accelerating computation. However,…

Machine Learning · Computer Science 2026-03-25 Hanglin Li , Shuchang Tian , Chen Lin , Zhiyong Zhao , Kun Zhan

Efficient deployment of large language models (LLMs) necessitates low-bit quantization to minimize model size and inference cost. While low-bit integer formats (e.g., INT8/INT4) have been the conventional choice, emerging low-bit…

Machine Learning · Computer Science 2023-05-23 Yijia Zhang , Lingran Zhao , Shijie Cao , Wenqiang Wang , Ting Cao , Fan Yang , Mao Yang , Shanghang Zhang , Ningyi Xu

Large Language Models (LLMs) today are powerful problem solvers across many domains, and they continue to get stronger as they scale in model size, training set size, and training set quality, as shown by extensive research and…

This note shares some simple calculations and experiments related to absmax-based blockwise quantization, as used in Dettmers et al., 2023. Their proposed NF4 data type is said to be information theoretically optimal for representing…

Machine Learning · Computer Science 2023-06-16 Davis Yoshida

Diffusion transformers have demonstrated remarkable capabilities in generating videos. However, their practical deployment is severely constrained by high memory usage and computational cost. Post-Training Quantization provides a practical…

Computer Vision and Pattern Recognition · Computer Science 2026-03-20 Rundong Su , Jintao Zhang , Zhihang Yuan , Haojie Duanmu , Jianfei Chen , Jun Zhu

Quantization has emerged as a standard technique for accelerating inference for generative models by enabling faster low-precision computations and reduced memory transfers. Recently, GPU accelerators have added first-class support for…

We propose LLM-FP4 for quantizing both weights and activations in large language models (LLMs) down to 4-bit floating-point values, in a post-training manner. Existing post-training quantization (PTQ) solutions are primarily integer-based…

Computation and Language · Computer Science 2024-04-30 Shih-yang Liu , Zechun Liu , Xijie Huang , Pingcheng Dong , Kwang-Ting Cheng

Model quantization reduces the bit-width of weights and activations, improving memory efficiency and inference speed in diffusion models. However, achieving 4-bit quantization remains challenging. Existing methods, primarily based on…

Machine Learning · Computer Science 2025-05-29 Maosen Zhao , Pengtao Chen , Chong Yu , Yan Wen , Xudong Tan , Tao Chen

The recent hardware-accelerated microscaling 4-bit floating-point formats such as MXFP4 and NVFP4, supported on NVIDIA and AMD GPUs, promise to revolutionize large language model (LLM) inference. Yet, their practical benefits remain…

Modern AI hardware, such as Nvidia's Blackwell architecture, is increasingly embracing low-precision floating-point (FP) formats to handle the pervasive activation outliers in Large Language Models (LLMs). Despite this industry trend, a…

Machine Learning · Computer Science 2025-10-30 Mengzhao Chen , Meng Wu , Hui Jin , Zhihang Yuan , Jing Liu , Chaoyi Zhang , Yunshui Li , Jie Huang , Jin Ma , Zeyue Xue , Zhiheng Liu , Xingyan Bin , Ping Luo

The recently introduced NVFP4 format demonstrates remarkable performance and memory benefits for quantized large language model (LLM) inference. However, we observe two types of redundancy in NVFP4 encoding: (1) The FP4 element format…

Machine Learning · Computer Science 2026-02-03 Yuzong Chen , Xilai Dai , Jake Hyun , Chi-Chih Chang , Wonsuk Jang , Yuheng Wu , Thierry Tambe , Jae-sun Seo , Mohamed S. Abdelfattah

The growing computational demands of training large language models (LLMs) necessitate more efficient methods. Quantized training presents a promising solution by enabling low-bit arithmetic operations to reduce these costs. While FP8…

Machine Learning · Computer Science 2026-05-18 Ruizhe Wang , Yeyun Gong , Xiao Liu , Guoshuai Zhao , Ziyue Yang , Baining Guo , Zhengjun Zha , Peng Cheng

Microscaling data formats leverage per-block tensor quantization to enable aggressive model compression with limited loss in accuracy. Unlocking their potential for efficient training and inference necessitates hardware-friendly…

Machine Learning · Computer Science 2026-01-28 Andrea Fasoli , Monodeep Kar , Chi-Chun Liu , Swagath Venkataramani , Viji Srinivasan , Leland Chang , Naigang Wang
‹ Prev 1 2 3 10 Next ›