Saved in:
Bibliographic Details
Main Authors: Lim, Gordon, Larson, Stefan, Leach, Kevin
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.13140
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910749931077632
author Lim, Gordon
Larson, Stefan
Leach, Kevin
author_facet Lim, Gordon
Larson, Stefan
Leach, Kevin
contents Tobacco3482 is a widely used document classification benchmark dataset. However, our manual inspection of the entire dataset uncovers widespread ontological issues, especially large amounts of annotation label problems in the dataset. We establish data label guidelines and find that 11.7% of the dataset is improperly annotated and should either have an unknown label or a corrected label, and 16.7% of samples in the dataset have multiple valid labels. We then analyze the mistakes of a top-performing model and find that 35% of the model's mistakes can be directly attributed to these label issues, highlighting the inherent problems with using a noisily labeled dataset as a benchmark. Supplementary material, including dataset annotations and code, is available at https://github.com/gordon-lim/tobacco3482-mistakes/.
format Preprint
id arxiv_https___arxiv_org_abs_2412_13140
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Label Errors in the Tobacco3482 Dataset
Lim, Gordon
Larson, Stefan
Leach, Kevin
Computer Vision and Pattern Recognition
Tobacco3482 is a widely used document classification benchmark dataset. However, our manual inspection of the entire dataset uncovers widespread ontological issues, especially large amounts of annotation label problems in the dataset. We establish data label guidelines and find that 11.7% of the dataset is improperly annotated and should either have an unknown label or a corrected label, and 16.7% of samples in the dataset have multiple valid labels. We then analyze the mistakes of a top-performing model and find that 35% of the model's mistakes can be directly attributed to these label issues, highlighting the inherent problems with using a noisily labeled dataset as a benchmark. Supplementary material, including dataset annotations and code, is available at https://github.com/gordon-lim/tobacco3482-mistakes/.
title Label Errors in the Tobacco3482 Dataset
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2412.13140