List your company
ResearchFormerly WizWand

Uncertainty Quantification on SciQ (test)

74.5AUROCSENTSAR
56.71661.33365.9570.567Jul 3, 2023
Updated 8mo ago

Evaluation results

MethodLinks
SENTSARLLM Model=Vicuna-13b, Number of generations=5, Rouge-L threshold=0.52023.0774.5
SARLLM Model=Vicuna-13b, Number of generations=5, Rouge-L threshold=0.52023.0774.1
SENTSARLLM Model=LLaMA-2-13b-chat, Number of generations=5, Rouge-L threshold=0.52023.0773.7
SARLLM Model=LLaMA-2-13b-chat, Number of generations=5, Rouge-L threshold=0.52023.0772.5
SENTSARLLM Model=WizardLM-13b, Number of generations=5, Rouge-L threshold=0.52023.0771.9
PELLM Model=LLaMA-2-13b-chat, Number of generations=5, Rouge-L threshold=0.52023.0771.8
TOKENSARLLM Model=LLaMA-2-13b-chat, Number of generations=5, Rouge-L threshold=0.52023.0771.8
SENTSARLLM Model=Vicuna-33b, Number of generations=5, Rouge-L threshold=0.52023.0771.7
SARLLM Model=Vicuna-33b, Number of generations=5, Rouge-L threshold=0.52023.0771
PELLM Model=Vicuna-13b, Number of generations=5, Rouge-L threshold=0.52023.0770.8
SARLLM Model=WizardLM-13b, Number of generations=5, Rouge-L threshold=0.52023.0770.7
TOKENSARLLM Model=Vicuna-13b, Number of generations=5, Rouge-L threshold=0.52023.0770.6
SELLM Model=LLaMA-2-13b-chat, Number of generations=5, Rouge-L threshold=0.52023.0769.2
LN-PELLM Model=LLaMA-2-13b-chat, Number of generations=5, Rouge-L threshold=0.52023.0768.8
TOKENSARLLM Model=WizardLM-13b, Number of generations=5, Rouge-L threshold=0.52023.0768.1
PELLM Model=WizardLM-13b, Number of generations=5, Rouge-L threshold=0.52023.0767.7
SELLM Model=Vicuna-13b, Number of generations=5, Rouge-L threshold=0.52023.0767.5
SELLM Model=Vicuna-33b, Number of generations=5, Rouge-L threshold=0.52023.0767.4
LN-PELLM Model=Vicuna-13b, Number of generations=5, Rouge-L threshold=0.52023.0766.8
LN-PELLM Model=Vicuna-33b, Number of generations=5, Rouge-L threshold=0.52023.0766.8
PELLM Model=Vicuna-33b, Number of generations=5, Rouge-L threshold=0.52023.0766.5
TOKENSARLLM Model=Vicuna-33b, Number of generations=5, Rouge-L threshold=0.52023.0766.5
SELLM Model=WizardLM-13b, Number of generations=5, Rouge-L threshold=0.52023.0764.9
LN-PELLM Model=WizardLM-13b, Number of generations=5, Rouge-L threshold=0.52023.0763.8
LSLLM Model=Vicuna-13b, Number of generations=5, Rouge-L threshold=0.52023.0758.9
LSLLM Model=Vicuna-33b, Number of generations=5, Rouge-L threshold=0.52023.0758.4
LSLLM Model=LLaMA-2-13b-chat, Number of generations=5, Rouge-L threshold=0.52023.0757.8
LSLLM Model=WizardLM-13b, Number of generations=5, Rouge-L threshold=0.52023.0757.4