paper-with-me

Papers

From a Lossless (~1.5:1) Compression Algorithm for Llama2 7B Weights to Variable Precision, Variable Range, Compressed Numeric Data Types for CNNs and LLMs

2024-04-16 · Vincenzo Liguori

This paper starts with a simple lossless ~1.5:1 compression algorithm for the weights of the Large Language Model (LLM) Llama2 7B [1] that can be implemented in ~200 LUTs in AMD FPGAs, processing over 800 million bfloat16 numbers per second. This framework is then extended to variable precision, variable range, compressed numerical data types that are a user defined super set of both floats and posits [2]. The paper then discusses a simple hardware implementation of such format based on ANS (Asymmetrical Numeral Systems) [3] that acts as a bridge between this flexible data format and a computational engine while, at the same time, achieving bandwidth reduction. An example of a token factory using weight compression and sharing is also given.

📄 PDF Abstract BibTeX arXiv:2404.10896

Code (0)

등록된 구현이 없습니다.

Tasks

Language ModelingLanguage ModellingLarge Language Model

Methods 이 논문이 사용한 방법론

SET Dynamic Sparse Training method where weight mask is updated randomly periodically

Similar Papers 제목 키워드 기반

ZipNN: Lossless Compression for AI Models

2024-11-07 · Moshik Hershcovitch, Andrew Wood, Leshem Choshen, Guy Girmonsky 외

With the growth of model sizes and the scale of their deployment, their sheer size burdens the infrastructure requiring more network and more storage to accommodate these. While there is a vast model compression literatu…

Model Compression

Llamazip: Leveraging LLaMA for Lossless Text Compression and Training Dataset Detection

2025-11-16 · Sören Dréano, Derek Molloy, Noel Murphy arxiv

This work introduces Llamazip, a novel lossless text compression algorithm based on the predictive capabilities of the LLaMA3 language model. Llamazip achieves significant data reduction by only storing tokens that the m…

70% Size, 100% Accuracy: Lossless LLM Compression for Efficient GPU Inference via Dynamic-Length Float

2025-04-15 · Tianyi Zhang, Yang Sui, Shaochen Zhong, Vipin Chaudhary 외

Large Language Models (LLMs) have grown rapidly in size, creating significant challenges for efficient deployment on resource-constrained hardware. In this paper, we introduce Dynamic-Length Float (DFloat11), a lossless …

CPUGPU

LLMZip: Lossless Text Compression using Large Language Models

2023-06-06 · Chandra Shekhara Kaushik Valmeekam, Krishna Narayanan, Dileep Kalathil, Jean-Francois Chamberland 외

We provide new estimates of an asymptotic upper bound on the entropy of English using the large language model LLaMA-7B as a predictor for the next token given a window of past tokens. This estimate is significantly smal…

Language ModelingLanguage ModellingLarge Language ModelText Compression

SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression

2023-06-05 · Tim Dettmers, Ruslan Svirschevski, Vage Egiazarian, Denis Kuznedelev 외

Recent advances in large language model (LLM) pretraining have led to high-quality LLMs with impressive abilities. By compressing such LLMs via quantization to 3-4 bits per parameter, they can fit into memory-limited dev…

GPULanguage ModellingLarge Language ModelQuantization