English
Related papers

Related papers: Entropy Law: The Story Behind Data Compression and…

200 papers

Data is a critical asset for training large language models (LLMs), alongside compute resources and skilled workers. While some training data is publicly available, substantial investment is required to generate proprietary datasets, such…

Machine Learning · Computer Science 2026-01-27 Mélissa Tamine , Otmane Sakhi , Benjamin Heymann

Language models demonstrate remarkable abilities when pre-trained on large text corpora and fine-tuned for specific tasks, but how and why pre-training shapes the success of the final model remains poorly understood. Notably, although…

Learning and compression are driven by the common aim of identifying and exploiting statistical regularities in data, which opens the door for fertile collaboration between these areas. A promising group of compression techniques for…

Machine Learning · Computer Science 2021-02-02 Fernando E. Rosas , Pedro A. M. Mediano , Michael Gastpar

Decoding strategies play a central role in shaping the reasoning ability of large language models (LLMs). Traditional methods such as greedy decoding and beam search often suffer from error propagation, while sampling-based approaches…

Natural language data follows a power-law distribution, with most knowledge and skills appearing at very low frequency. While a common intuition suggests that reweighting or curating data towards a uniform distribution may help models…

Artificial Intelligence · Computer Science 2026-04-28 Zixuan Wang , Xingyu Dang , Jason D. Lee , Kaifeng Lyu

Image Coding for Machines (ICM) is becoming more important as research in computer vision progresses. ICM is a vital research field that pursues the use of images for image recognition models, facilitating efficient image transmission and…

Computer Vision and Pattern Recognition · Computer Science 2024-10-17 Takahiro Shindo , Taiju Watanabe , Yui Tatsumi , Hiroshi Watanabe

Large Language Models (LLMs) need to adapt to the continuous changes in data, tasks, and user preferences. Due to their massive size and the high costs associated with training, LLMs are not suitable for frequent retraining. However,…

Computation and Language · Computer Science 2024-12-11 Dongfang Li , Zetian Sun , Xinshuo Hu , Baotian Hu , Min Zhang

Deep learning is increasingly viewed as a dynamical process in parameter space, yet many existing theories still treat training as a closed optimization system. This view is limited for real-world AI, where models operate under uncertainty,…

Machine Learning · Computer Science 2026-05-25 Kim Phuc Tran

Data compression continues to evolve, with traditional information theory methods being widely used for compressing text, images, and videos. Recently, there has been growing interest in leveraging Generative AI for predictive compression…

Information Theory · Computer Science 2024-09-24 Swathi Shree Narashiman , Nitin Chandrachoodan

Large Language Models (LLMs) achieve strong performance across tasks, but face storage and compute challenges on edge devices. We propose EntroLLM, a compression framework combining mixed quantization and entropy coding to reduce storage…

Machine Learning · Computer Science 2026-05-05 Arnab Sanyal , Gourav Datta , Prithwish Mukherjee , Sandeep P. Chinchali , Michael Orshansky

Large language models (LLMs) rely on pretraining on massive and heterogeneous corpora, where training data composition has a decisive impact on training efficiency and downstream generalization under realistic compute and data budget…

Computation and Language · Computer Science 2026-04-21 Zhuo Chen , Yuxuan Miao , Supryadi , Deyi Xiong

In large language models (LLMs), each block operates on the residual stream to map input token sequences to output token distributions. However, most of the interpretability literature focuses on internal latent representations, leaving…

Machine Learning · Computer Science 2026-02-03 Riccardo Ali , Francesco Caso , Christopher Irwin , Pietro Liò

Guided by the belief of the scaling law, large language models (LLMs) have achieved impressive performance in recent years. However, scaling law only gives a qualitative estimation of loss, which is influenced by various factors such as…

Computation and Language · Computer Science 2024-09-16 Chuhan Wu , Ruiming Tang

We consider the computational aspects of lossy data compression problem, where the compression error is determined by a cover of the data space. We propose an algorithm which reduces the number of partitions needed to find the entropy with…

Information Theory · Computer Science 2012-04-03 Marek Śmieja , Jacek Tabor

We study the problem of entropy calibration, which asks whether a language model's entropy over generations matches its log loss on human text. Past work found that models are miscalibrated, with entropy per step increasing as generations…

Computation and Language · Computer Science 2026-01-14 Steven Cao , Gregory Valiant , Percy Liang

Code large language models (Code LLMs) are powerful but costly to train, with scaling laws predicting performance from model size, data, and compute. However, different programming languages (PLs) have varying impacts during pre-training…

Computation and Language · Computer Science 2025-12-16 Jian Yang , Shawn Guo , Lin Jing , Wei Zhang , Aishan Liu , Chuan Hao , Zhoujun Li , Wayne Xin Zhao , Xianglong Liu , Weifeng Lv , Bryan Dai

Large language models (LLMs) risk retaining unauthorized or sensitive information from their training data, which raises privacy concerns. LLM unlearning seeks to mitigate these risks by selectively removing specified data while maintaining…

Computation and Language · Computer Science 2025-05-29 Hwan Chang , Hwanhee Lee

Scaling laws have shaped recent advances in machine learning by enabling predictable scaling of model performance based on model size, computation, and data volume. Concurrently, the rise in computational cost for AI has motivated model…

Machine Learning · Computer Science 2025-06-03 Andrei Panferov , Alexandra Volkova , Ionut-Vlad Modoranu , Vage Egiazarian , Mher Safaryan , Dan Alistarh

The principle of maximum entropy is a broadly applicable technique for computing a distribution with the least amount of information possible constrained to match empirical data, for instance, feature expectations. We seek to generalize…

Information Theory · Computer Science 2022-05-30 Kenneth Bogert

Communication has emerged as a critical bottleneck in the distributed training of large language models (LLMs). While numerous approaches have been proposed to reduce communication overhead, the potential of lossless compression has…

Distributed, Parallel, and Cluster Computing · Computer Science 2026-05-01 Wenxiang Lin , Xinglin Pan , Ruibo Fan , Shaohuai Shi , Xiaowen Chu