List your company

Question Answering on QASC (R@1, R@3, N@1, N@3)

74.17Recall@1S-DPO
65.787667.963870.1472.3162May 11, 2026

Evaluation results

MethodLinks
S-DPOModel=Llama32026.0574.1792.3874.1785.13
DPO-kModel=Llama32026.0573.9592.3873.9584.93
MASS-DPOModel=Llama32026.0573.8492.673.8485.13
MASS-DPOModel=SmolLM32026.0572.5291.8372.5283.82
MASS-DPOModel=Qwen32026.0571.8591.6171.8583.73
DPOModel=Llama32026.0571.391.7271.383.41
DPO-kModel=Qwen32026.0571.0892.7271.0883.95
S-DPOModel=SmolLM32026.0570.5391.3970.5382.8
S-DPOModel=Qwen32026.0570.4291.570.4283.04
DPO-kModel=SmolLM32026.0570.3190.1870.3182.06
DMPOModel=Llama32026.0569.3291.9469.3282.67
DMPOModel=Qwen32026.0568.2190.5168.2181.4
DPOModel=Qwen32026.0567.7790.6267.7781.29
DPOModel=SmolLM32026.0567.1190.0767.1180.7
DMPOModel=SmolLM32026.0566.1188.7466.1179.54