Epstein Files: OCR Text from Unsealed Court Documents

A page-level OCR dataset compiled from a release of documents related to the Jeffrey Epstein case, published by author 'ishumilin' on Hugging Face. The dataset represents scanned PDF documents processed through a proprietary automated OCR pipeline. The dataset page was last updated on March 19, 2026.

Use Cases

• Text retrieval tasks based on the OCR output of court documents.
• Text classification tasks based on the content of the released files.
• Question answering tasks based on the information extracted from the scanned pages.

Strengths

• Dataset is structured at the page level for individual document analysis.
• Significantly expands upon an earlier release of datasets 1-8.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count, file formats, and license are unknown, which may limit suitability assessment.

Provenance

Source

Documents related to the Jeffrey Epstein case.

Collection Method

Compiled from scanned PDFs using a proprietary automated OCR pipeline.

Time Range

null

Freshness

Last updated 2026-03-19 13:09:09; freshness should be verified.

Geography

null