List your company
ResearchFormerly WizWand

Sentiment Classification on Yelp5 (test)

98.5AccuracyICL-gold
16.8638.05559.2580.445Dec 19, 2022
Updated 8mo ago

Evaluation results

MethodLinks
ICL-goldPrompting Method=Direct, Backbone=GPT-3, Oracle=true2022.1298.5
Z-ICLPrompting Method=Direct, Backbone=GPT-32022.1297.7
No-demosPrompting Method=Direct, Backbone=GPT-32022.1296.4
ICL-randomPrompting Method=Direct, Backbone=GPT-3, Oracle=true2022.1293.7
ICL-goldPrompting Method=Channel, Backbone=GPT-3, Oracle=true2022.1291.7
ICL-randomPrompting Method=Channel, Backbone=GPT-3, Oracle=true2022.1290.4
No-demosPrompting Method=Channel, Backbone=GPT-32022.1288
Z-ICLPrompting Method=Channel, Backbone=GPT-32022.1287.6
Majority2022.1250
ICL-randomModel=GPT-J, Prompting=Channel, Training data access=true2022.1248
ICL-goldModel=GPT-J, Prompting=Direct, Training data access=true2022.1247.5
ICL-goldModel=GPT-J, Prompting=Channel, Training data access=true2022.1247.4
Z-ICLModel=GPT-J, Prompting=Channel, Training data access=false2022.1244.2
ICL-randomModel=GPT-J, Prompting=Direct, Training data access=true2022.1243.5
ICL-goldModel=GPT-NeoX, Prompting=Channel, Training data access=true2022.1243.5
ICL-randomModel=GPT-NeoX, Prompting=Channel, Training data access=true2022.1243.5
Naive Z-ICLModel=GPT-J, Prompting=Channel, Training data access=false2022.1241.8
ICL-randomModel=GPT-NeoX, Prompting=Direct, Training data access=true2022.1241.3
Naive Z-ICLModel=GPT-J, Prompting=Direct, Training data access=false2022.1241.2
Z-ICLModel=GPT-J, Prompting=Direct, Training data access=false2022.1240.8
ICL-goldModel=GPT-NeoX, Prompting=Direct, Training data access=true2022.1240.6
Random inputsModel=GPT-J, Prompting=Channel, Training data access=false2022.1240.5
Z-ICLModel=GPT-NeoX, Prompting=Channel, Training data access=false2022.1239.9
Naive Z-ICLModel=GPT-NeoX, Prompting=Direct, Training data access=false2022.1239.1
Z-ICLModel=GPT-NeoX, Prompting=Direct, Training data access=false2022.1238.6
Random inputsModel=GPT-NeoX, Prompting=Direct, Training data access=false2022.1237.1
Naive Z-ICLModel=GPT-NeoX, Prompting=Channel, Training data access=false2022.1236.9
No-demosModel=GPT-J, Prompting=Channel, Training data access=false2022.1236.6
Random inputsModel=GPT-J, Prompting=Direct, Training data access=false2022.1236.4
Random inputsModel=GPT-NeoX, Prompting=Channel, Training data access=false2022.1229.1
No-demosModel=GPT-J, Prompting=Direct, Training data access=false2022.1228.7
No-demosModel=GPT-NeoX, Prompting=Channel, Training data access=false2022.1228.6
No-demosModel=GPT-NeoX, Prompting=Direct, Training data access=false2022.1221.3
MajorityTraining data access=false2022.1220