RoBERTa Baseline

Reported on 7 benchmarks across 5 tasks

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing5 results

Sentiment AnalysisonDPM
F1-score
49.1
Sentiment AnalysisonDPM
Macro-F1
10.4
SemEval-2022 Task 4-1 (Binary PCL Detection)onDPM
F1-score
49.1
Binary Condescension DetectiononDPM
F1-score
49.1
best: 63.69 (BERT-PCL)
Multi-label Condescension DetectiononDPM
Macro-F1
10.4
best: 44.4 (Prompt)

Music2 results

Pcl DetectiononDPM
F1-score
49.1
Pcl DetectiononDPM
Macro-F1
10.4