TweetEval
Benchmarks
TaskDatasetSOTA resultTrendResults
TweetEval
72.69Accuracy
112
TweetEVAL (test)
84.17Accuracy (A)
44
TweetEval hate
67.53Macro F1
21
TweetEval 1.0 (test)
34.2Emoji (M-F1)
18
TweetEval Emotion
68.74Macro-F1
15
TweetEval latest (test)
0.297Emoji
9
TweetEval (IID)
53.1Accuracy
8
TweetEval Offensive (test)
69.14Accuracy
8
TweetEval Hate (test)
55.72Accuracy
8
TweetEval T-Sentiment (test)
1.23AU-MSE
8
TweetEval T-Hate (test)
4.88AU-MSE
8
TweetEval T-Emotions (test)
1.47AU-MSE
8
TweetEval irony (test)
84.18Accuracy
7
TweetEval offensive
68.3Macro F1
6
TweetEval irony
62.7Macro F1
6
TweetEval stance-feminist (test)
41.3Macro F1
6
TweetEval
72F1
3
TweetEval stance-atheism
27.4Macro F1
3
TweetEval stance-atheism (TW-A) (test)
0.285Macro-F1
3
TweetEval-offensive (Tw-O) (test)
56.3Macro F1
3
TweetEval-hate (Tw-H) (test)
59Macro F1
3