AceParse: A Comprehensive Dataset with Diverse Structured Texts for Academic Literature Parsing

2024-09-16

arXiv.org

AI Summary

Academic PDF parsing suffers from scarce fine-grained annotations for structured elements—such as mathematical formulas, tables, lists, algorithms, and inline mathematical expressions—leading to poor model generalization. Method: We introduce the first benchmark dataset covering all five complex structural categories, accompanied by a unified, fine-grained annotation schema. Leveraging this dataset, we propose AceParser, a multimodal Transformer that jointly processes PDF text and rendered page images, and simultaneously optimizes structure-aware sequence labeling and graph generation tasks. Contribution/Results: AceParser achieves state-of-the-art performance, outperforming prior methods by +4.1% in F1 score and +5.0% in Jaccard similarity. Both the benchmark dataset and the AceParser model are publicly released to enable reproducible research in academic document structure parsing.

Technology Category

• Machine Learning: Structured Learning
• Natural Language Processing: Syntax — Tagging, Chunking & Parsing
• Computer Vision: Multi-modal Vision

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Web Mining and Content Analysis: Bridging structured and unstructured data
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

With the development of data-centric AI, the focus has shifted from model-driven approaches to improving data quality. Academic literature, as one of the crucial types, is predominantly stored in PDF formats and needs to be parsed into texts before further processing. However, parsing diverse structured texts in academic literature remains challenging due to the lack of datasets that cover various text structures. In this paper, we introduce AceParse, the first comprehensive dataset designed to support the parsing of a wide range of structured texts, including formulas, tables, lists, algorithms, and sentences with embedded mathematical expressions. Based on AceParse, we fine-tuned a multimodal model, named AceParser, which accurately parses various structured texts within academic literature. This model outperforms the previous state-of-the-art by 4.1% in terms of F1 score and by 5% in Jaccard Similarity, demonstrating the potential of multimodal models in academic literature parsing. Our dataset is available at https://github.com/JHW5981/AceParse.

Problem

• PDF Parsing
• AI Training Data
• Complex Document Format

Innovation

• AceParse Dataset
• AceParser Model
• Enhanced Parsing Accuracy