Document-level Machine Translation on EN-ZH standard (test)
23BLEUL2-C-7B
Evaluation results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| L2-C-7BCategory=Chat Models2026.01 | 23 | 23 | 22 | 23 | 22 | 23 | 20 | |
| L2-C-70BCategory=Chat Models2026.01 | 22 | 17 | 23 | 22 | 23 | 22 | 21 | |
| NLLB-54BCategory=NMT Baselines2026.01 | 21 | 20 | 20 | 18 | 19 | 19 | 13 | |
| NLLB-600MCategory=NMT Baselines2026.01 | 20 | 19 | 19 | 19 | 18 | 18 | 11 | |
| DS-671BCategory=Reasoning2026.01 | 20 | 15 | 21 | 21 | 20 | 20 | 22 | |
| NLLB-3.3BCategory=NMT Baselines2026.01 | 19 | 18 | 18 | 16 | 17 | 17 | 15 | |
| DS-Qwen-7BCategory=Reasoning2026.01 | 18 | 14 | 17 | 17 | 15 | 16 | 16 | |
| Llama2-7BCategory=Pre-trained LLMs2026.01 | 17 | 16 | 15 | 15 | 16 | 15 | 8 | |
| mBART-50Category=NMT Baselines2026.01 | 16 | 15 | 16 | 13 | 14 | 14 | 17 | |
| DS-Llama-8BCategory=Reasoning2026.01 | 15 | 13 | 14 | 14 | 13 | 12 | 18 | |
| L3-C-8BCategory=Chat Models2026.01 | 14 | 12 | 13 | 11 | 11 | 11 | 12 | |
| Llama3-8BCategory=Pre-trained LLMs2026.01 | 13 | 11 | 12 | 10 | 10 | 10 | 10 | |
| Q2.5-C-7BCategory=Chat Models2026.01 | 11 | 8 | 8 | 9 | 8 | 9 | 9 | |
| Q3-R-8BCategory=Reasoning2026.01 | 10 | 10 | 11 | 12 | 12 | 13 | 19 | |
| Llama2-70BCategory=Pre-trained LLMs2026.01 | 9 | 7 | 7 | 7 | 7 | 8 | 7 | |
| Q3-C-8BCategory=Chat Models2026.01 | 8 | 6 | 6 | 6 | 6 | 6 | 5 | |
| L3-C-70BCategory=Chat Models2026.01 | 7 | 9 | 5 | 5 | 5 | 3 | 2 | |
| MiniCPMCategory=Reasoning2026.01 | 6 | 9 | 10 | 8 | 9 | 7 | 14 | |
| Qwen2.5-7BCategory=Pre-trained LLMs2026.01 | 5 | 5 | 4 | 4 | 4 | 5 | 1 | |
| Llama3-70BCategory=Pre-trained LLMs2026.01 | 4 | 4 | 3 | 3 | 3 | 4 | 4 | |
| Q2.5-C-72BCategory=Chat Models2026.01 | 3 | 3 | 2 | 2 | 2 | 2 | 6 | |
| Qwen2.5-72BCategory=Pre-trained LLMs2026.01 | 1 | 1 | 1 | 1 | 1 | 1 | 3 |