Lateral Thinking on BRAINTEASER Sentence Puzzle SemEval 2024 Task 9 (test)
92.5Original ScoreDeBERTaV3 + AMMC + train-data-wp+sp + Humor + RiddleSense
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| DeBERTaV3 + AMMC + train-data-wp+sp + Humor + RiddleSenseBackbone=DeBERTaV3, Architecture=AutoModelForMultipleChoice (AMMC), Training Data=wp+sp, Data Augmentation=Humor + RiddleSense2024.05 | 92.5 | 95 | 90 | 92.5 | 87.5 | 92.5 | |
| Human2024.05 | 90.7 | 90.7 | 94.4 | 90.7 | 88.9 | 92 | |
| DeBERTaV3 + AMMC + train-data-wp+spBackbone=DeBERTaV3, Architecture=AutoModelForMultipleChoice (AMMC), Training Data=wp+sp2024.05 | 90 | 90 | 85 | 90 | 82.5 | 88.3 | |
| BERT-base + AMMC + train-data-wp+spBackbone=BERT-base, Architecture=AutoModelForMultipleChoice (AMMC), Training Data=wp+sp2024.05 | 65 | 62.5 | 62.5 | 60 | 50 | 60 | |
| ChatGPT2024.05 | 60.8 | 59.3 | 67.9 | 50.7 | 39.7 | 62.7 | |
| BERT-base + AMSC + train-data-wp+spBackbone=BERT-base, Architecture=AutoModelForSequenceClassification (AMSC), Training Data=wp+sp2024.05 | 47.5 | 55 | 50 | 35 | 25 | 50.8 | |
| RoBERTa-LBackbone=RoBERTa-Large2024.05 | 43.5 | 40.2 | 46.4 | 33 | 20.1 | 43.4 |