Human Preference Evaluation for Code-switched Text Generation on EN-CS (Out of domain)
434.5ScoreGold Standard
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Gold Standard2025.02 | 434.5 | 1 | |
| Llama3training=fine-tuned2025.02 | 282 | 2 | |
| NLLBtraining=fine-tuned, backbone=nllb-200-distilled-600M2025.02 | 247.5 | 3 | |
| Llama3 Instructtraining=fine-tuned2025.02 | 210 | 4 | |
| Llama3.3-70Bfsmode=5-shot2025.02 | 164 | 5 | |
| GPT-4ofsmode=5-shot2025.02 | 162 | 6 |