Optical Character Recognition2020en
mldoc-zero-shot-english-to-italian
Dataset from Papers With Code
Metrics:accuracy, cer, wer, f1
Current State of the Art
MultiFiT, pseudo
Unknown
76.02
accuracy
accuracy Progress Over Time
Showing 3 breakthroughs from May 2018 to Sep 2019
Key Milestones
May 2018
MultiCCA + CNN
From paper: A Corpus for Multilingual Document Classification in Eight Languages
69.4
Dec 2018
Massively Multilingual Sentence Embeddings
From paper: Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond
69.4
+0.1%
Sep 2019
MultiFiT, pseudoCurrent SOTA
From paper: MultiFiT: Efficient Multi-lingual Language Model Fine-tuning
76.0
+9.5%
Total Improvement
9.6%
Time Span
1y 4m
Breakthroughs
3
Current SOTA
76.0
Top Models Performance Comparison
Top 4 models ranked by accuracy
Best Score
76.0
Top Model
MultiFiT, pseudo
Models Compared
4
Score Range
15.3
accuracyPrimary
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | MultiFiT, pseudo | 76.02 | Sep 2019 | |
| 2 | Massively Multilingual Sentence Embeddings | 69.43 | Dec 2018 | |
| 3 | MultiCCA + CNN | 69.38 | May 2018 | |
| 4 | BiLSTM (Europarl) | 60.73 | May 2018 |
Related Papers3
MultiFiT: Efficient Multi-lingual Language Model Fine-tuning
Sep 2019Models: MultiFiT, pseudo
Massively Multilingual Sentence Embeddings for Zero-Shot Cross-Lingual Transfer and Beyond
Dec 2018Models: Massively Multilingual Sentence Embeddings
A Corpus for Multilingual Document Classification in Eight Languages
May 2018Models: MultiCCA + CNN, BiLSTM (Europarl)