List your company

Uncertainty-guided selective generation on CoQA (dev)

96.52AUARC (Token Entropy)UA-CLM
83.093686.579390.06593.5507Dec 3, 2024

Evaluation results

MethodLinks
UA-CLMModel=Gemma-2B2024.1296.5296.6896.7196.72
UA-CLMModel=Llama-2-13B2024.1296.459797.8497.92
UA-CLMModel=Llama-2-7B2024.1296.0396.5796.9997.16
CLMModel=Llama-2-13B2024.1295.7995.8496.5996.61
ULTModel=Llama-2-13B2024.1295.195.3495.9296.12
CLMModel=Gemma-2B2024.1294.6894.5496.1996.55
CLMModel=Llama-2-7B2024.1294.3594.4495.0895.3
Pre-trainedModel=Gemma-2B2024.1292.7193.3992.3594.52
ULTModel=Llama-2-7B2024.1292.1292.1993.1593.26
ULTModel=Gemma-2B2024.1291.7291.5290.9391.68
Pre-trainedModel=Llama-2-13B2024.1286.7289.491.3792.09
CTModel=Llama-2-7B2024.1286.0387.890.2990.75
CTModel=Llama-2-13B2024.1284.9786.4789.4492
Pre-trainedModel=Llama-2-7B2024.1283.6186.0693.4894.11