This dataset contains outputs from davanstrien/dataset-paper-pages-smoke processed with NuExtract3, a 4B vision-language model for document understanding.

Processing Details

Source Dataset: davanstrien/dataset-paper-pages-smoke Model: numind/NuExtract3 Mode: structured-extraction Number of Samples: 230 Processing Time: 5.6 min Processing Date: 2026-06-01 10:20 UTC

Configuration

Image Column:… See the full description on the dataset page: https://huggingface.co/datasets/davanstrien/dataset-paper-pages-smoke-routed.