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

Sentiment Classification research benchmarks – Page 9 · SOTA2 Research

Sentiment Classification

Benchmarks

Dataset NameSOTA MethodSort

Dataset NameSOTA methodMetricTrendResultsLast Updated

0BVE (Avg Acc Diff)

4

Feb 26, 2026

Amazon Books (Amzbook) (test)

0Avg Explanation Impact on Accuracy

4

Feb 26, 2026

IMDB 30% symmetric noise (test)

88.01Accuracy

4

Feb 26, 2026

74.91F1 Score

4

Feb 26, 2026

75.97F1 Score

4

Feb 26, 2026

IMDB-review heteroskedastic simulated 5%/40% noise (val)

94.3Accuracy (Negative Class)

4

Feb 26, 2026

Yelp Extreme (test)

0.121Loss

4

Feb 26, 2026

Yelp Bulk (test)

0.097Classification Loss

4

Feb 26, 2026

Amazon Overall (test)

0.084Classification Loss

4

Feb 26, 2026

Amazon Extreme (test)

0.08Loss

4

Feb 26, 2026

Amazon Bulk (test)

0.085Loss

4

Feb 26, 2026

Imbalanced IMDB (val)

2.86Error Rate (Positive)

4

Feb 26, 2026

SST SentEval (test)

84.6Accuracy

4

Feb 26, 2026

CR SentEval (test)

87Accuracy

4

Feb 26, 2026

IMDB character-level Long Range Arena (LRA) (val)

70.19Accuracy

3

Apr 13, 2026

94.6Original Accuracy

3

Feb 26, 2026

0.87Accuracy

3

Feb 26, 2026

86.61Accuracy

3

Feb 26, 2026

90.4Accuracy

2

May 7, 2026

87.4Accuracy

2

May 7, 2026

86.7Accuracy

2

May 7, 2026

80.8Accuracy

2

May 7, 2026

78.5Accuracy

2

May 7, 2026

Sherlock Holmes MEG dataset (test)

144.588t-statistic

2

Feb 26, 2026

SFT Llama3 8B Instruct

0.95Accuracy

2

Feb 26, 2026

Page 9 of 10

Popular datasets

IMDB SST Yelp Amazon Amazon Review

Follow for updates

Join SOTA2 Contact