Optical Character Recognition2020en
reuters-rcv1/rcv2-german-to-english
Dataset from Papers With Code
Metrics:accuracy, cer, wer, f1
Current State of the Art
Biinclusion (Euro500kReuters)
Unknown
84.4
accuracy
accuracy Progress Over Time
Showing 3 breakthroughs from Dec 2013 to Dec 2014
Key Milestones
Dec 2014
Biinclusion (Euro500kReuters)Current SOTA
From paper: Leveraging Monolingual Data for Crosslingual Compositional Word Representations
84.4
+6.6%
Total Improvement
9.8%
Time Span
1y
Breakthroughs
3
Current SOTA
84.4
Top Models Performance Comparison
Top 3 models ranked by accuracy
Best Score
84.4
Top Model
Biinclusion (Euro...
Models Compared
3
Score Range
7.5
accuracyPrimary
| # | Model | Score | Paper / Code | Date |
|---|---|---|---|---|
| 1 | Biinclusion (Euro500kReuters) | 84.4 | Dec 2014 | |
| 2 | Bi+ | 79.2 | Apr 2014 | |
| 3 | biCVM+ | 76.9 | Dec 2013 |
Related Papers3
Leveraging Monolingual Data for Crosslingual Compositional Word Representations
Dec 2014Models: Biinclusion (Euro500kReuters)
Multilingual Models for Compositional Distributed Semantics
Apr 2014Models: Bi+
Multilingual Distributed Representations without Word Alignment
Dec 2013Models: biCVM+