Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering on TweetQA (eval)
Loading...
80.81
Clean F1
SDBN-p
66.146
69.953
73.76
77.567
Jun 9, 2026
Clean F1
Delete-Char F1
Delete-Word F1
Updated 1mo ago
Evaluation Results
Method
Method
Links
Clean F1
Delete-Char F1
Delete-Word F1
SDBN-p
Backbone=LLaMA-2-7B, P...
2026.06
80.81
65.55
64.15
FreeLB
Backbone=LLaMA-2-7B, P...
2026.06
76.57
60.59
60.79
EDA
Backbone=LLaMA-2-7B, P...
2026.06
70.02
53.14
55.04
NEFTune
Backbone=LLaMA-2-7B, P...
2026.06
69.34
55.13
54.3
Vanilla
Backbone=LLaMA-2-7B, P...
2026.06
68.09
51.56
56.06
SMART
Backbone=LLaMA-2-7B, P...
2026.06
66.71
49.54
51.13
Feedback
Search any
task
Search any
task