Codesota · Benchmark · tobacco-3482Home/Leaderboards/tobacco-3482
Unknown

tobacco-3482.

tobacco-3482 is a state-of-the-art machine learning benchmark indexed on Codesota. This page tracks published model results, top scores per metric, and the SOTA timeline for tobacco-3482.

Paper ↗Leaderboard ↓
§ 01 · SOTA history

Year over year.

§ 02 · Leaderboard

Results by metric.

Found a wrong score or missing run?
Use row edits to send a sourced correction into moderation.
Add / edit result ↗Report issue ↗

Accuracy

Accuracy is the reported evaluation metric for tobacco-3482. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Accuracyverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksFix
01HEADoC-Large
HEADoC LARGE (90.58M params) hybrid multimodal model combining visual and textual features via deep attention. Published in Progress in Artificial Intelligence, Springer, 2025.
verified96.662025Source ↗Looks wrong?
02HEADoC-Base
HEADoC BASE (27.7M params) compact hybrid multimodal model, most parameter-efficient SOTA model. Published in Progress in Artificial Intelligence, Springer, 2025.
verified95.982025Source ↗Looks wrong?
03DocXClassifier-FPN
DocXClassifierFPN variant with Feature Pyramid Networks. Published in IJDAR vol. 27, pp. 447-473, June 2024. Top-1 accuracy with transfer learning from RVL-CDIP.
verified95.712024Source ↗Looks wrong?
04DocXClassifier-L
From paper: DocXClassifier: High Performance Explainable Deep Network for Document Image Classification
verified95.572022Paper ↗Code ↗Looks wrong?
05Multimodal Side-Tuning (MobileNetV2)
From paper: Multimodal Side-Tuning for Document Classification
verified90.52023Paper ↗Code ↗Looks wrong?
06Multimodal Side-Tuning (ResNet50)
From paper: Multimodal Side-Tuning for Document Classification
verified90.32023Paper ↗Code ↗Looks wrong?
07DiT-Base
DiT (Document Image Transformer) pre-trained on RVL-CDIP weights, evaluated via 4-fold cross-validation on Tobacco3482. Adjusted accuracy 89.7% when accounting for dataset label errors. Reported in label error analysis paper, Dec 2024.
verified84.12024Source ↗Looks wrong?
08DocBert [DOCBERT]
From paper: Efficient Document Image Classification Using Region-Based Graph Neural Network
verified82.32021Paper ↗Looks wrong?
09BERT [BERT]
From paper: Efficient Document Image Classification Using Region-Based Graph Neural Network
verified792021Paper ↗Looks wrong?
10Eff-GNN + Word2Vec [word2vec] + Image Embedding
From paper: Efficient Document Image Classification Using Region-Based Graph Neural Network
verified77.52021Paper ↗Looks wrong?
11Eff-GNN + Word2Vec [word2vec]
From paper: Efficient Document Image Classification Using Region-Based Graph Neural Network
verified73.52021Paper ↗Looks wrong?
12Optimized Text CNN
From paper: Light-Weighted CNN for Text Classification
verified462020Paper ↗Code ↗Looks wrong?
13Lightweight TextCNN with Dual Optimizer
From paper: Light-Weighted CNN for Text Classification
verified43.52020Paper ↗Code ↗Looks wrong?
14Lightweight Text CNN
From paper: Light-Weighted CNN for Text Classification
verified422020Paper ↗Code ↗Looks wrong?

Memory

Memory is the reported evaluation metric for tobacco-3482. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Memoryverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksFix
01VGG
From paper: Efficient Document Image Classification Using Region-Based Graph Neural Network
verified7.082021Paper ↗Looks wrong?

Training Time Hours

Training Time Hours is the reported evaluation metric for tobacco-3482. Codesota tracks published model scores on this metric so readers can compare state-of-the-art results across sources and model families.

Higher is better

Trust tiers for Training Time Hoursverifiedpapervendorcommunityunverified
RankModelTrustScoreYearLinksFix
01Optimized Text CNN
From paper: Light-Weighted CNN for Text Classification
verified2.002020Paper ↗Code ↗Looks wrong?
02Lightweight Text CNN
From paper: Light-Weighted CNN for Text Classification
verified1.002020Paper ↗Code ↗Looks wrong?
03Lightweight TextCNN with Dual Optimizer
From paper: Light-Weighted CNN for Text Classification
verified0.432020Paper ↗Code ↗Looks wrong?
§ 04 · Submit a result

Add to the leaderboard.

← Back to Leaderboards