List your company

Closed-book Question Answering on SimpleQA (train)

18.8AccuracySFT
-0.2324.7099.6514.591May 16, 2026

Evaluation results

MethodLinks
SFTModel=Qwen3-4B-It2026.0518.8
SFTModel=Qwen3-8B2026.0516.8
MIXSDModel=Qwen3-4B-It, lambda=0.32026.0510.5
MIXSDModel=Qwen3-4B-It, lambda=0.52026.0510.4
MIXSDModel=Qwen3-8B, lambda=0.52026.059.3
MIXSDModel=Qwen3-8B, lambda=0.32026.059
SFTModel=Qwen3-1.7B2026.058.5
OPSDModel=Qwen3-4B-It, n=1, T=02026.058.5
OPSDModel=Qwen3-8B2026.058.3
OPSD-NLLModel=Qwen3-4B-It, n=1, T=02026.057.8
OPSDModel=Qwen3-4B-It2026.057.5
MIXSDModel=Qwen3-1.7B, lambda=0.32026.057.1
OPSDModel=Qwen3-8B, n=1, T=02026.056.8
MIXSDModel=Qwen3-1.7B, lambda=0.52026.056.3
OPSD-NLLModel=Qwen3-8B, n=1, T=02026.056
OPSDModel=Qwen3-1.7B, n=1, T=02026.053.9
OPSD-NLLModel=Qwen3-1.7B, n=1, T=02026.053.4
BaseModel=Qwen3-4B-It2026.053.4
OPSDModel=Qwen3-1.7B2026.052.9
BaseModel=Qwen3-8B2026.052.5
BaseModel=Qwen3-1.7B2026.050.5