Natural Language Understanding on NusaX
81.37Macro F1FLARE MT
Evaluation results
| Method | Links | |
|---|---|---|
| FLARE MTTransfer Strategy=Translate-Train, Base Model=mT5-XL2025.01 | 81.37 | |
| FLARETransfer Strategy=Translate-Train, Base Model=Gemma 2 9B2025.01 | 80.86 | |
| FLARETransfer Strategy=Translate-Train, Base Model=mT5-XL2025.01 | 80.54 | |
| LoRATransfer Strategy=Translate-Train, Base Model=mT5-XL2025.01 | 80.41 | |
| input-level fusionTransfer Strategy=Translate-Train, Base Model=mT5-XL2025.01 | 80.12 | |
| X-MixupTransfer Strategy=Translate-Train, Base Model=Gemma 2 9B2025.01 | 79.94 | |
| LoRATransfer Strategy=Translate-Train, Base Model=Gemma 2 9B2025.01 | 79.37 | |
| FLARETransfer Strategy=Translate-Train, Base Model=XLM-R Large2025.01 | 79.18 | |
| X-MixupTransfer Strategy=Translate-Train, Base Model=mT5-XL2025.01 | 78.61 | |
| input-level fusionTransfer Strategy=Translate-Train, Base Model=XLM-R Large2025.01 | 78.53 | |
| FLARE MTTransfer Strategy=Translate-Train, Base Model=Gemma 2 9B2025.01 | 78.09 | |
| input-level fusionTransfer Strategy=Translate-Train, Base Model=Gemma 2 9B2025.01 | 77.98 | |
| FLARE MTTransfer Strategy=Translate-Train, Base Model=XLM-R Large2025.01 | 77.18 | |
| LoRATransfer Strategy=Translate-Train, Base Model=XLM-R Large2025.01 | 77 | |
| X-MixupTransfer Strategy=Translate-Train, Base Model=XLM-R Large2025.01 | 76.37 | |
| FLARETransfer Strategy=Translate-Train, Base Model=Llama 3.1 8B2025.01 | 76.08 | |
| X-MixupTransfer Strategy=Translate-Train, Base Model=Llama 3.1 8B2025.01 | 75.91 | |
| mT5-XLTransfer Strategy=Translate-Test2025.01 | 75.77 | |
| XLM-R LargeTransfer Strategy=Zero-Shot Cross-Lingual Transfer2025.01 | 75.26 | |
| XLM-R LargeTransfer Strategy=Translate-Test2025.01 | 74.85 | |
| LoRATransfer Strategy=Translate-Train, Base Model=Llama 3.1 8B2025.01 | 74.76 | |
| mT5-XLTransfer Strategy=Zero-Shot Cross-Lingual Transfer2025.01 | 74.72 | |
| input-level fusionTransfer Strategy=Translate-Train, Base Model=Llama 3.1 8B2025.01 | 74.6 | |
| FLARE MTTransfer Strategy=Translate-Train, Base Model=Llama 3.1 8B2025.01 | 74.52 | |
| Llama 3.1 8BTransfer Strategy=Translate-Test2025.01 | 72.67 | |
| Llama 3.1 8BTransfer Strategy=Zero-Shot Cross-Lingual Transfer2025.01 | 71.74 | |
| Gemma 2 9BTransfer Strategy=Zero-Shot Cross-Lingual Transfer2025.01 | 71.61 | |
| Gemma 2 9BTransfer Strategy=Translate-Test2025.01 | 71.61 |