Sentence-type prediction on AdMIRe multilingual (15 languages) 2.0 (test)
59AccuracyEnsemble (Qwen3+GPT-4o)
Evaluation Results
| Method | Links | |
|---|---|---|
| Ensemble (Qwen3+GPT-4o)System architecture=Qwen3 + GPT-4o ensemble2026.02 | 59 | |
| GPT-4o2026.02 | 58 | |
| Qwen3-32B2026.02 | 56.5 | |
| Llama3.1-70B2026.02 | 55 | |
| Mistral-large2026.02 | 54.5 | |
| DeepSeek-v32026.02 | 53.5 | |
| CLIP baselineSystem architecture=XLM-RoBERTa-ViT-H-142026.02 | 53.3 |