List your company

Multiple-choice task on ECARE KRE

82.38AccuracyIRCAN-CAD
46.208855.599464.9974.3806Jun 26, 2024

Evaluation results

MethodLinks
IRCAN-CADBackbone=Gemma-2B-it2024.0682.3817.62
IRCAN-CADBackbone=LLaMA-2-7B-Chat2024.0680.9619.04
IRCAN-CADBackbone=LLaMA-2-13B-Chat2024.0678.4221.58
CADBackbone=LLaMA-2-13B-Chat2024.0678.1321.87
CADBackbone=Gemma-2B-it2024.0677.7622.24
IRCANBackbone=Gemma-2B-it2024.0677.3822.62
Based_on_FormattedBackbone=Gemma-2B-it2024.0676.7223.28
Utilizing_FormattedBackbone=Gemma-2B-it2024.0676.4423.56
IRCANBackbone=LLaMA-2-7B-Chat2024.0676.0623.94
Based_onBackbone=Gemma-2B-it2024.0675.5924.41
OriginalBackbone=Gemma-2B-it2024.0675.4924.51
ITI (Mass Mean Shift)Backbone=Gemma-2B-it2024.0673.826.2
OpinBackbone=LLaMA-3-8B-Instruct2024.0673.826.2
CADBackbone=LLaMA-2-7B-Chat2024.0673.723.3
ITI (Probe Weight Direction)Backbone=LLaMA-2-13B-Chat2024.0673.4226.58
ITI (Probe Weight Direction)Backbone=Gemma-2B-it2024.0673.0426.96
ITI (Mass Mean Shift)Backbone=LLaMA-2-13B-Chat2024.0671.4428.56
IRCANBackbone=LLaMA-2-13B-Chat2024.0669.8430.16
CADBackbone=LLaMA-3-8B-Instruct2024.0669.5630.44
IRCAN-CADBackbone=LLaMA-3-8B-Instruct2024.0669.3730.63
Based_on_FormattedBackbone=LLaMA-3-8B-Instruct2024.0669.0930.91
Utilizing_FormattedBackbone=LLaMA-3-8B-Instruct2024.0668.9931.01
OriginalBackbone=LLaMA-2-13B-Chat2024.0668.931.1
Based_on_FormattedBackbone=LLaMA-2-13B-Chat2024.0668.6131.39
Based_onBackbone=LLaMA-2-13B-Chat2024.0667.8632.14
Utilizing_FormattedBackbone=LLaMA-2-13B-Chat2024.0666.1633.84
OpinBackbone=Gemma-2B-it2024.0663.5236.48
Based_on_FormattedBackbone=LLaMA-2-7B-Chat2024.0663.1536.85
Based_onBackbone=LLaMA-2-7B-Chat2024.0661.5538.45
Utilizing_FormattedBackbone=LLaMA-2-7B-Chat2024.0660.9839.02
ITI (Probe Weight Direction)Backbone=LLaMA-3-8B-Instruct2024.0660.5139.49
Based_onBackbone=LLaMA-3-8B-Instruct2024.0659.140.9
IRCANBackbone=LLaMA-3-8B-Instruct2024.0657.8742.13
OriginalBackbone=LLaMA-3-8B-Instruct2024.0657.442.6
OpinBackbone=LLaMA-2-13B-Chat2024.0657.1242.88
OriginalBackbone=LLaMA-2-7B-Chat2024.0655.0444.96
OpinBackbone=LLaMA-2-7B-Chat2024.0655.0444.96
ITI (Probe Weight Direction)Backbone=LLaMA-2-7B-Chat2024.0649.5850.42
ITI (Mass Mean Shift)Backbone=LLaMA-3-8B-Instruct2024.0649.5850.42
ITI (Mass Mean Shift)Backbone=LLaMA-2-7B-Chat2024.0647.652.4