Saved in:
Bibliographic Details
Main Author: Wan, Jun
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2501.06802
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909485026508800
author Wan, Jun
author_facet Wan, Jun
contents In 2020, OpenAI proposed the first type of Scaling Laws, describing the relationships between model loss and the scale of parameters, data, and training computation. In 2024, OpenAI proposed the second type of Scaling Laws, describing the relationship between model inference performance and inference computation. In this paper, we analyze LLMs training and inference processes from the perspective of lossless compression using conditional Kolmogorov complexity, and unify these two types of Scaling Laws. We find that both types of Scaling Laws improve approximation of conditional Kolmogorov complexity by increasing execution steps of Turing machine. The first type of Scaling Laws increases execution steps by increasing number of model parameters. The second type of Scaling Laws increases execution steps by increasing the number of intermediate tokens.
format Preprint
id arxiv_https___arxiv_org_abs_2501_06802
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unifying Two Types of Scaling Laws from the Perspective of Conditional Kolmogorov Complexity
Wan, Jun
Artificial Intelligence
In 2020, OpenAI proposed the first type of Scaling Laws, describing the relationships between model loss and the scale of parameters, data, and training computation. In 2024, OpenAI proposed the second type of Scaling Laws, describing the relationship between model inference performance and inference computation. In this paper, we analyze LLMs training and inference processes from the perspective of lossless compression using conditional Kolmogorov complexity, and unify these two types of Scaling Laws. We find that both types of Scaling Laws improve approximation of conditional Kolmogorov complexity by increasing execution steps of Turing machine. The first type of Scaling Laws increases execution steps by increasing number of model parameters. The second type of Scaling Laws increases execution steps by increasing the number of intermediate tokens.
title Unifying Two Types of Scaling Laws from the Perspective of Conditional Kolmogorov Complexity
topic Artificial Intelligence
url https://arxiv.org/abs/2501.06802