Machine Translation on De-En document-level
88.45d-COMETOurs
Evaluation Results
| Method | Links | |
|---|---|---|
| OursBase Model=Mistral-Nemo-Instruct, Quality-aware fine-tuning=true2025.04 | 88.45 | |
| OursBase Model=LLaMA-3-8B-Instruct, Quality-aware fine-tuning=true2025.04 | 88.14 | |
| DocRefine (doc)Base Model=Mistral-Nemo-Instruct2025.04 | 88.13 | |
| DocRefine (sent)Base Model=Mistral-Nemo-Instruct2025.04 | 88.09 | |
| Ours (-QA)Base Model=LLaMA-3-8B-Instruct, Quality-aware fine-tuning=false2025.04 | 88.02 | |
| Ours (-QA)Base Model=Mistral-Nemo-Instruct, Quality-aware fine-tuning=false2025.04 | 88.01 | |
| Sent2Sent (tuned)Base Model=LLaMA-3-8B-Instruct2025.04 | 87.94 | |
| Doc2Doc (tuned)Base Model=LLaMA-3-8B-Instruct2025.04 | 87.82 | |
| Doc2Doc (tuned)Base Model=Mistral-Nemo-Instruct2025.04 | 87.8 | |
| SentRefine (sent)Base Model=Mistral-Nemo-Instruct2025.04 | 87.73 | |
| DocRefine (doc)Base Model=LLaMA-3-8B-Instruct2025.04 | 87.71 | |
| Doc2DocBase Model=Mistral-Nemo-Instruct2025.04 | 87.61 | |
| DocRefine (sent)Base Model=LLaMA-3-8B-Instruct2025.04 | 87.42 | |
| Doc2DocBase Model=LLaMA-3-8B-Instruct2025.04 | 87.05 | |
| OursBase Model=Mistral-Nemo-Instruct, Quality-aware fine-tuning=true2025.04 | 86.89 | |
| Sent2Sent (tuned)Base Model=Mistral-Nemo-Instruct2025.04 | 86.86 | |
| Sent2SentBase Model=Mistral-Nemo-Instruct2025.04 | 86.85 | |
| Ours (-QA)Base Model=Mistral-Nemo-Instruct, Quality-aware fine-tuning=false2025.04 | 86.7 | |
| DocRefine (sent)Base Model=Mistral-Nemo-Instruct2025.04 | 86.58 | |
| Doc2Doc (tuned)Base Model=Mistral-Nemo-Instruct2025.04 | 86.5 | |
| DocRefine (doc)Base Model=Mistral-Nemo-Instruct2025.04 | 86.44 | |
| SentRefine (sent)Base Model=Mistral-Nemo-Instruct2025.04 | 86.36 | |
| OursBase Model=LLaMA-3-8B-Instruct, Quality-aware fine-tuning=true2025.04 | 86.05 | |
| Sent2SentBase Model=LLaMA-3-8B-Instruct2025.04 | 85.97 | |
| Ours (-QA)Base Model=LLaMA-3-8B-Instruct, Quality-aware fine-tuning=false2025.04 | 85.7 | |
| Sent2Sent (tuned)Base Model=Mistral-Nemo-Instruct2025.04 | 85.49 | |
| Doc2Doc (tuned)Base Model=LLaMA-3-8B-Instruct2025.04 | 85.45 | |
| Sent2Sent (tuned)Base Model=LLaMA-3-8B-Instruct2025.04 | 85.42 | |
| DocRefine (sent)Base Model=LLaMA-3-8B-Instruct2025.04 | 85.38 | |
| SentRefine (sent)Base Model=LLaMA-3-8B-Instruct2025.04 | 85.21 | |
| DocRefine (doc)Base Model=LLaMA-3-8B-Instruct2025.04 | 85.07 | |
| Sent2SentBase Model=Mistral-Nemo-Instruct2025.04 | 84.66 | |
| Doc2DocBase Model=Mistral-Nemo-Instruct2025.04 | 84.34 | |
| SentRefine (sent)Base Model=LLaMA-3-8B-Instruct2025.04 | 83.7 | |
| Doc2DocBase Model=LLaMA-3-8B-Instruct2025.04 | 83.35 | |
| Sent2SentBase Model=LLaMA-3-8B-Instruct2025.04 | 82.5 |