List your company

Generation correctness prediction on SciQ

77.99AUROCParaGrad
56.035661.735367.43573.1347May 6, 2026

Evaluation results

MethodLinks
ParaGradModel=Mistral-Nemo-Instruct12B2026.0577.99
ExGradModel=Mistral-Nemo-Instruct12B2026.0577.53
HybridGradModel=Mistral-Nemo-Instruct12B2026.0576.9
G-NLLModel=Mistral-Nemo-Instruct12B2026.0576.83
LN-PEModel=Mistral-Nemo-Instruct12B2026.0576.68
SARModel=Mistral-Nemo-Instruct12B2026.0576.57
HybridGradModel=Llama3.1-Instruct8B2026.0576.31
SemGradModel=Llama3.1-Instruct8B2026.0575.76
SemGradModel=Mistral-Nemo-Instruct12B2026.0575.55
G-NLLModel=Llama3.1-Instruct8B2026.0575.49
SARModel=Llama3.1-Instruct8B2026.0575.28
ParaGradModel=Llama3.1-Instruct8B2026.0574.98
DegModel=Mistral-Nemo-Instruct12B2026.0574.15
ExGradModel=Llama3.1-Instruct8B2026.0574.11
S.D.Model=Llama3.1-Instruct8B2026.0574
M.I.Model=Mistral-Nemo-Instruct12B2026.0573.72
DegModel=Llama3.1-Instruct8B2026.0573.11
HybridGradModel=Qwen3-Instruct4B2026.0572.83
SARModel=Qwen3-Instruct4B2026.0572.72
G-NLLModel=Qwen3-Instruct4B2026.0572.7
S.D.Model=Mistral-Nemo-Instruct12B2026.0572.52
LN-PEModel=Llama3.1-Instruct8B2026.0572.51
M.I.Model=Llama3.1-Instruct8B2026.0572.43
SemGradModel=Qwen3-Instruct4B2026.0572.2
ParaGradModel=Qwen3-Instruct4B2026.0572.09
INSIDEModel=Mistral-Nemo-Instruct12B2026.0571.54
Self-ConModel=Llama3.1-Instruct8B2026.0571.47
P(True)Model=Mistral-Nemo-Instruct12B2026.0571.4
ExGradModel=Qwen3-Instruct4B2026.0571.34
Self-ConModel=Mistral-Nemo-Instruct12B2026.0571.07
INSIDEModel=Llama3.1-Instruct8B2026.0570.83
S.E.Model=Llama3.1-Instruct8B2026.0570.27
S.E.Model=Mistral-Nemo-Instruct12B2026.0568.53
LN-PEModel=Qwen3-Instruct4B2026.0567.08
M.I.Model=Qwen3-Instruct4B2026.0566.25
DegModel=Qwen3-Instruct4B2026.0565.01
P(True)Model=Llama3.1-Instruct8B2026.0564.91
S.D.Model=Qwen3-Instruct4B2026.0563.79
Self-ConModel=Qwen3-Instruct4B2026.0561.95
INSIDEModel=Qwen3-Instruct4B2026.0557.96
P(True)Model=Qwen3-Instruct4B2026.0557.13
S.E.Model=Qwen3-Instruct4B2026.0556.88