Boundary detection on Hybrid essays Out-of-Domain (test)
43.6#Bry=1 F1 ScoreTriBERT
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| TriBERTPrototype size (p)=4, Encoder fine-tuning status=Fine-tuned2023.07 | 43.6 | 55.1 | 42.8 | 47.7 | |
| TriBERTPrototype size (p)=5, Encoder fine-tuning status=Fine-tuned2023.07 | 42.4 | 51.1 | 40.2 | 45.2 | |
| TriBERTPrototype size (p)=3, Encoder fine-tuning status=Fine-tuned2023.07 | 42.1 | 56.7 | 46.3 | 48.4 | |
| TriBERTPrototype size (p)=6, Encoder fine-tuning status=Fine-tuned2023.07 | 41.9 | 48.7 | 39.5 | 43.9 | |
| TriBERTPrototype size (p)=2, Encoder fine-tuning status=Fine-tuned2023.07 | 41.7 | 59.7 | 54.5 | 51 | |
| TriBERTPrototype size (p)=1, Encoder fine-tuning status=Fine-tuned2023.07 | 37.9 | 55.9 | 64 | 49.7 | |
| BERTEncoder fine-tuning status=Fine-tuned2023.07 | 36.9 | 54.5 | 63.2 | 48.6 | |
| TriBERT (NT)Prototype size (p)=3, Encoder fine-tuning status=Not Trained2023.07 | 20.6 | 30.5 | 28.8 | 25.9 | |
| TriBERT (NT)Prototype size (p)=2, Encoder fine-tuning status=Not Trained2023.07 | 20.5 | 31.6 | 30.2 | 26.6 | |
| TriBERT (NT)Prototype size (p)=4, Encoder fine-tuning status=Not Trained2023.07 | 20.1 | 29.2 | 26.5 | 24.8 | |
| TriBERT (NT)Prototype size (p)=5, Encoder fine-tuning status=Not Trained2023.07 | 19.1 | 28.7 | 26.2 | 24 | |
| TriBERT (NT)Prototype size (p)=6, Encoder fine-tuning status=Not Trained2023.07 | 18.9 | 27.6 | 24.9 | 23.3 | |
| TriBERT (NT)Prototype size (p)=1, Encoder fine-tuning status=Not Trained2023.07 | 18.8 | 27.8 | 30.6 | 24.4 | |
| GPTZero2023.07 | 16.3 | 22.4 | 24.1 | 20.2 | |
| LREncoder fine-tuning status=Fine-tuned2023.07 | 15.9 | 24.8 | 24.1 | 20.8 | |
| RANDOM2023.07 | 13 | 20.4 | 20.9 | 17.3 |