Question Answering on IndoCulture
81.4AccuracySailor2-L-20B-Chat
Evaluation Results
| Method | Links | |
|---|---|---|
| Sailor2-L-20B-ChatRAG=IndoSoSci + Wikipedia, D=20, Zero-shot=true2026.01 | 81.4 | |
| Qwen-SEA-LION-v4-32B-ITRAG=IndoSoSci + Wikipedia, D=20, Zero-shot=true2026.01 | 79.5 | |
| Sailor2-L-20BRAG=IndoSoSci + Wikipedia, D=20, Zero-shot=true2026.01 | 78 | |
| Sailor2-L-8B-ChatRAG=IndoSoSci + Wikipedia, D=20, Zero-shot=true2026.01 | 77.2 | |
| Sailor2-L-20B-ChatRAG=None, Zero-shot=true2026.01 | 75.4 | |
| Sailor2-L-8BRAG=IndoSoSci + Wikipedia, D=20, Zero-shot=true2026.01 | 74.3 | |
| Sailor2-L-20BRAG=None, Zero-shot=true2026.01 | 72.1 | |
| Qwen-SEA-LION-v4-32B-ITRAG=None, Zero-shot=true2026.01 | 70.9 | |
| Sailor2-L-8B-ChatRAG=None, Zero-shot=true2026.01 | 70.5 | |
| SEALLMs-v3-7B-ChatRAG=IndoSoSci + Wikipedia, D=20, Zero-shot=true2026.01 | 66.3 | |
| Sailor2-L-8BRAG=None, Zero-shot=true2026.01 | 64.2 | |
| SEALLMs-v3-7BRAG=IndoSoSci + Wikipedia, D=20, Zero-shot=true2026.01 | 64.1 | |
| SEALLMs-v3-7B-ChatRAG=None, Zero-shot=true2026.01 | 60.6 | |
| SEALLMs-v3-7BRAG=None, Zero-shot=true2026.01 | 54.6 |