List your company

AlpacaEval 2.0 on UltraMedical Preference

17.7LCSSPO
1.585.7659.9514.135Oct 28, 2025

Evaluation results

MethodLinks
SSPOBackbone=Llama3 (8B), Size=10%2025.1017.718.4
SPABackbone=Llama3 (8B), Size=10%2025.1015.716.9
KTOBackbone=Llama3 (8B), Size=10%2025.1014.215.2
SSRMBackbone=Mistral (7B), Size=10%2025.1013.114.6
SSRMBackbone=Llama3 (8B), Size=10%2025.1012.915.8
SSPOBackbone=Mistral (7B), Size=10%2025.101217
SPABackbone=Mistral (7B), Size=10%2025.1011.316.2
DPOBackbone=Mistral (7B), Size=10%2025.1011.27.2
KTOBackbone=Mistral (7B), Size=10%2025.1011.115.6
SimPOBackbone=Llama3 (8B), Size=10%2025.1010.54.5
SimPOBackbone=Mistral (7B), Size=10%2025.1010.27.6
DPOBackbone=Mistral (7B), Size=1%2025.108.74.8
DPOBackbone=Llama3 (8B), Size=10%2025.107.86.2
SSPOBackbone=Mistral (7B), Size=1%2025.107.613.2
ORPOBackbone=Llama3 (8B), Size=10%2025.107.55.6
SPABackbone=Mistral (7B), Size=1%2025.106.710.1
ORPOBackbone=Mistral (7B), Size=10%2025.106.55
KTOBackbone=Mistral (7B), Size=1%2025.106.49.8
SimPOBackbone=Mistral (7B), Size=1%2025.1067.2
SSRMBackbone=Mistral (7B), Size=1%2025.1065.9
SSPOBackbone=Llama3 (8B), Size=1%2025.105.16.7
SPABackbone=Llama3 (8B), Size=1%2025.104.35.8
SSRMBackbone=Llama3 (8B), Size=1%2025.103.95
ORPOBackbone=Mistral (7B), Size=1%2025.103.73.1
KTOBackbone=Llama3 (8B), Size=1%2025.103.75.4
SimPOBackbone=Llama3 (8B), Size=1%2025.102.83.8
DPOBackbone=Llama3 (8B), Size=1%2025.102.65.3
ORPOBackbone=Llama3 (8B), Size=1%2025.102.24.8