Skip to content

Dataset · Careerflow

Benchmark for resume parsing

Careerflow/ResumeBench

Benchmark for evaluating the accuracy of structured data extraction from resumes by JSON Schema.

What's inside

The dataset contains 300 real, handwritten, and adversarial English PDF resumes with annotations. It is designed for testing document analysis systems at the field level.

What you can do with it

  • Evaluating accuracy of document parsers and data extraction systems
  • Developing automated candidate screening tools
Tasks
Document question answering, Embeddings
Languages
English
Size
up to 1K rows
Downloads this month
1.9K
Likes
2
Access
on request on Hugging Face

How to load it

from datasets import load_dataset

ds = load_dataset("Careerflow/ResumeBench")