Categories Companies Research

Loading the SOTA2 catalog…

Explore

Categories Companies Research

Contribute

Join SOTA2 Contact

SOTA2

© 2026 SOTA2

Build in Bellevue, USA

WizWand is now SOTA2 Research

Papers Benchmarks Datasets Tasks

Document Classification research benchmarks – Page 2 · SOTA2 Research

Document Classification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

92.5Accuracy

10

Feb 26, 2026

95.17Accuracy

10

Feb 26, 2026

95.69Accuracy

10

Feb 26, 2026

Rakuten Amazon JA

83.34Accuracy

8

May 6, 2026

Amazon Review ZH

InfoCTM + LLM-XTM

73.21Accuracy

8

May 6, 2026

77.85Accuracy

8

May 6, 2026

79.75Accuracy

8

May 6, 2026

Embedding_CPE + MLP

71.28Macro F1

8

Feb 26, 2026

EURLEX57K (test)

73.32Micro F1

8

Feb 26, 2026

48.1English

7

Feb 26, 2026

TED corpus L2 -> en (test)

45.2Arabic

7

Feb 26, 2026

TED corpus en -> L2 (test)

42.9Arabic

7

Feb 26, 2026

EURLEX57K Inverted (test)

72.44Micro F1

7

Feb 26, 2026

64.94Accuracy

7

Feb 26, 2026

SW-attention (max-bound)

92.2Accuracy

6

Mar 4, 2026

20 Newsgroups (held-out)

55F1

6

Feb 26, 2026

AG-News (held-out)

87F1 Score

6

Feb 26, 2026

Document Classification (DC) (test)

97.64Accuracy

6

Feb 26, 2026

Document Classification (DC) (dev)

0.977Accuracy

6

Feb 26, 2026

79.8Micro F1

5

Feb 26, 2026

llm-prune-5pointcontext

47.65Recall

5

Feb 26, 2026

Clinical-Longformer

84.2Accuracy

5

Feb 26, 2026

Clinical-Longformer

0.762AUC

5

Feb 26, 2026

Clinical-Longformer

97.7AUC

5

Feb 26, 2026

67.24Accuracy

5

Feb 26, 2026

Page 2 of 4

Popular datasets

Yelp LUN HP EURLEX Tobacco

Follow for updates

Join SOTA2 Contact