OpenReview Dataset: Academic Papers Converted to Markdown for LLM Reviewing

Academic papers were collected from OpenReview using its official API and converted into Markdown format using Nougat OCR. This dataset was developed as part of a course project for CSCI-GA 2565: Machine Learning at NYU. The last recorded update was on 2025-05-12.

Use Cases

• Training large language models to perform structured academic paper reviewing based on paper content.
• Developing text cleaning and preprocessing pipelines for academic PDFs converted to Markdown.
• Benchmarking OCR and text extraction quality from academic PDFs.
• Studying the structure and content of papers submitted to the OpenReview platform.

Strengths

• Papers were collected using the official OpenReview API, suggesting a direct source connection.
• A custom cleaning pipeline involving regular expressions was applied to the Markdown text.
• The dataset was created for a specific machine learning course project at NYU, providing a clear development context.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license information are unknown, which may limit suitability assessment.
• The description is incomplete and redirects to an external page for full details.

Provenance

Source

OpenReview platform

Collection Method

Collected via the official OpenReview API, with PDFs converted to Markdown using Nougat OCR.

Freshness

Last updated 2025-05-12 18:16:40.