MT-Bench on DSP Business
7MT ScoreSSPO
Evaluation Results
| Method | Links | |
|---|---|---|
| SSPOBackbone=Mistral (7B), Size=10%2025.10 | 7 | |
| SPABackbone=Mistral (7B), Size=10%2025.10 | 6.9 | |
| SSPOBackbone=Mistral (7B), Size=1%2025.10 | 6.9 | |
| DPOBackbone=Mistral (7B), Size=10%2025.10 | 6.8 | |
| ORPOBackbone=Mistral (7B), Size=1%2025.10 | 6.8 | |
| ORPOBackbone=Mistral (7B), Size=10%2025.10 | 6.8 | |
| SimPOBackbone=Mistral (7B), Size=10%2025.10 | 6.8 | |
| SSRMBackbone=Mistral (7B), Size=10%2025.10 | 6.8 | |
| KTOBackbone=Mistral (7B), Size=10%2025.10 | 6.8 | |
| DPOBackbone=Mistral (7B), Size=1%2025.10 | 6.7 | |
| SimPOBackbone=Mistral (7B), Size=1%2025.10 | 6.7 | |
| SPABackbone=Mistral (7B), Size=1%2025.10 | 6.7 | |
| SSRMBackbone=Mistral (7B), Size=1%2025.10 | 6.6 | |
| KTOBackbone=Mistral (7B), Size=1%2025.10 | 6.6 | |
| DPOBackbone=Llama3 (8B), Size=10%2025.10 | 5.7 | |
| SimPOBackbone=Llama3 (8B), Size=10%2025.10 | 5.7 | |
| KTOBackbone=Llama3 (8B), Size=10%2025.10 | 5.7 | |
| SPABackbone=Llama3 (8B), Size=10%2025.10 | 5.7 | |
| DPOBackbone=Llama3 (8B), Size=1%2025.10 | 5.6 | |
| ORPOBackbone=Llama3 (8B), Size=1%2025.10 | 5.6 | |
| ORPOBackbone=Llama3 (8B), Size=10%2025.10 | 5.6 | |
| SimPOBackbone=Llama3 (8B), Size=1%2025.10 | 5.6 | |
| SSRMBackbone=Llama3 (8B), Size=1%2025.10 | 5.6 | |
| SSRMBackbone=Llama3 (8B), Size=10%2025.10 | 5.6 | |
| KTOBackbone=Llama3 (8B), Size=1%2025.10 | 5.6 | |
| SPABackbone=Llama3 (8B), Size=1%2025.10 | 5.6 | |
| SSPOBackbone=Llama3 (8B), Size=1%2025.10 | 5.6 | |
| SSPOBackbone=Llama3 (8B), Size=10%2025.10 | 5.6 |