List your company

Retrieval-Augmented Generation on NQ (Reliability Score (RS))

54.33Reliability Score (RS)CDA
20.769229.482138.19546.9079Dec 17, 2024

Evaluation results

MethodLinks
CDABackbone=LLAMA3 8B INSTRUCT2024.1254.33
CDA-MBackbone=LLAMA3 8B INSTRUCT2024.1254.32
FSBBackbone=LLAMA3 8B INSTRUCT2024.1253.66
CDA-MBackbone=LLAMA2 13B CHAT2024.1251.95
ACD-ABackbone=LLAMA3 8B INSTRUCT2024.1251.8
CDABackbone=LLAMA2 13B CHAT2024.1251.42
ENTROPY (first-token)Backbone=LLAMA3 8B INSTRUCT2024.1250.84
ABSTAINBackbone=LLAMA3 8B INSTRUCT2024.1250.74
FSBBackbone=LLAMA2 13B CHAT2024.1249.88
CDA-MBackbone=MISTRAL 7B INSTRUCT2024.1249.51
ENTROPY (max)Backbone=LLAMA3 8B INSTRUCT2024.1249.4
CDABackbone=MISTRAL 7B INSTRUCT2024.1249.36
ABSTAINBackbone=MISTRAL 7B INSTRUCT2024.1249.11
FSBBackbone=MISTRAL 7B INSTRUCT2024.1249.05
ENTROPY (first-token)Backbone=MISTRAL 7B INSTRUCT2024.1248.65
ENTROPY (max)Backbone=LLAMA2 13B CHAT2024.1248.61
ENTROPY (average)Backbone=LLAMA3 8B INSTRUCT2024.1248.46
ACD-ABackbone=MISTRAL 7B INSTRUCT2024.1248.42
ENTROPY (max)Backbone=MISTRAL 7B INSTRUCT2024.1248.38
ACD-ABackbone=LLAMA2 13B CHAT2024.1248.3
ENTROPY (first-token)Backbone=LLAMA2 13B CHAT2024.1248.28
ENTROPY (average)Backbone=LLAMA2 13B CHAT2024.1247.63
CDABackbone=LLAMA2 7B CHAT2024.1247.09
ENTROPY (average)Backbone=MISTRAL 7B INSTRUCT2024.1246.84
CDA-MBackbone=LLAMA2 7B CHAT2024.1246.4
SELF-ASKBackbone=LLAMA3 8B INSTRUCT2024.1245.67
ENTROPY (min)Backbone=LLAMA3 8B INSTRUCT2024.1244.76
ENTROPY (min)Backbone=LLAMA2 13B CHAT2024.1243.73
ENTROPY (average)Backbone=LLAMA2 7B CHAT2024.1242.88
ENTROPY (min)Backbone=MISTRAL 7B INSTRUCT2024.1242.58
ENTROPY (first-token)Backbone=LLAMA2 7B CHAT2024.1241.94
ABSTAINBackbone=LLAMA2 13B CHAT2024.1241.54
ENTROPY (max)Backbone=LLAMA2 7B CHAT2024.1241.36
ENTROPY (min)Backbone=LLAMA2 7B CHAT2024.1241.3
SELF-ASKBackbone=MISTRAL 7B INSTRUCT2024.1239.34
ACDBackbone=LLAMA3 8B INSTRUCT2024.1238.4
FSBBackbone=LLAMA2 7B CHAT2024.1237.36
ACDBackbone=LLAMA2 13B CHAT2024.1235.65
SELF-ASKBackbone=LLAMA2 13B CHAT2024.1234.15
ACD-ABackbone=LLAMA2 7B CHAT2024.1233.56
CONTEXTBackbone=LLAMA3 8B INSTRUCT2024.1233.35
ACDBackbone=LLAMA2 7B CHAT2024.1233.02
ACDBackbone=MISTRAL 7B INSTRUCT2024.1232.42
SELF-ASKBackbone=LLAMA2 7B CHAT2024.1231.72
CONTEXTBackbone=LLAMA2 13B CHAT2024.1231.15
CONTEXTBackbone=MISTRAL 7B INSTRUCT2024.1230.68
CONTEXTBackbone=LLAMA2 7B CHAT2024.1230.11
CADBackbone=LLAMA3 8B INSTRUCT2024.1228.86
CADBackbone=MISTRAL 7B INSTRUCT2024.1228.6
CADBackbone=LLAMA2 13B CHAT2024.1228.4
CADBackbone=LLAMA2 7B CHAT2024.1227.2
ABSTAINBackbone=LLAMA2 7B CHAT2024.1222.06