Binary classification of sense ID on Fool me if you can (Set 4)
85.2F1 ScoreQwen 3 (Reasoning)*
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen 3 (Reasoning)*# Parameters=4B, fine-tuned=false2026.03 | 85.2 | |
| Gemma 3 (Reasoning)*# Parameters=4B, fine-tuned=false2026.03 | 81.1 | |
| FLAN-T5-xl# Parameters=3B2026.03 | 71.8 | |
| Roberta-base# Parameters=125M2026.03 | 71.5 | |
| Mixtral 7bx8# Parameters=7B2026.03 | 71.4 | |
| T5-xl# Parameters=3B2026.03 | 71 | |
| T5-large# Parameters=770M2026.03 | 69.1 | |
| Bert-large# Parameters=340M2026.03 | 68.9 | |
| Llama3 8B# Parameters=8B2026.03 | 68.7 | |
| FLAN-T5-large# Parameters=780M2026.03 | 66.3 |