Code as Representation: A Compilable Parsing Paradigm for Academic Documents

2026-08-18

AI Summary

本文提出了一种可编译的学术文档解析范式CADP，通过将页面重构为LaTeX和可执行Python代码来解决科学文档中结构化元素表示不佳的问题。

Abstract

Academic papers are a primary carrier of scientific knowledge, yet most of this knowledge remains locked in PDFs that are optimized for human reading rather than machine use. For Multimodal Large Language Models (MLLMs), the core challenge is not only perception, but representation: scientific pages interleave text with Structured Academic Elements (SAEs) such as tables, formulas, charts, and pseudocode, whose structure, data, and logic are poorly preserved by common surrogates like Markdown. We therefore propose Compilable Academic Document Parsing (CADP), a paradigm that reconstructs a full page as contextual \LaTeX{} plus executable Python, so that structure-preserving elements and executable chart representations can be reconstructed, recompiled, and directly verified against the source page. To support this setting, we introduce CADP-Bench, an expert-verified benchmark of full academic pages containing tightly coupled text and multiple SAE types, evaluated through a re-injection compilation protocol. We further study current capabilities using SOTA MLLMs and an exploratory multi-agent baseline that incorporates common agentic techniques. Results show that even frontier models still struggle to produce high-fidelity executable reconstructions, highlighting substantial room for improvement in structure-aware scientific document parsing. CADP-Bench is released for future research.

Problem

• Academic Documents
• Scientific Knowledge
• Structured Academic Elements
• Representation
• Multimodal Large Language Models

Innovation

• Compilable Academic Document Parsing (CADP)
• Structured Academic Elements (SAEs)
• executable representations
• CADP-Bench
• re-injection compilation