LaTeX Compilation: Challenges in the Era of LLMs

TL;DR

This paper critically examines TeX's limitations in the context of LLM-assisted scientific writing and introduces Mogan STEM, a WYSIWYG editor that outperforms TeX in efficiency, semantics, and tool integration, with benefits for LLM fine-tuning.

Contribution

It identifies fundamental defects in TeX and proposes Mogan STEM as a superior alternative with improved performance and lower entropy for LLM applications.

Findings

Mogan outperforms TeX in compilation and rendering speed.

Using .tmu format is more efficient for LLM fine-tuning than TeX.

Extensive experiments verify Mogan’s benefits in LLM tasks.

Abstract

As large language models (LLMs) increasingly assist scientific writing, limitations and the significant token cost of TeX become more and more visible. This paper analyzes TeX's fundamental defects in compilation and user experience design to illustrate its limitations on compilation efficiency, generated semantics, error localization, and tool ecosystem in the era of LLMs. As an alternative, Mogan STEM, a WYSIWYG structured editor, is introduced. Mogan outperforms TeX in the above aspects by its efficient data structure, fast rendering, and on-demand plugin loading. Extensive experiments are conducted to verify the benefits on compilation/rendering time and performance in LLM tasks. Furthermore, we show that due to Mogan's lower information entropy, it is more efficient to use .tmu (the document format of Mogan) to fine-tune LLMs than TeX. Therefore, we launch an appeal for larger experiments on LLM training using the .tmu format.