Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Claim-level Uncertainty Quantification on FactScore English (test)
Loading...
71
ROC-AUC
CCP
52.28
57.14
62
66.86
Mar 7, 2024
ROC-AUC
Updated 4d ago
Evaluation Results
Method
Method
Links
ROC-AUC
CCP
Base LLM=Jais 13b
2024.03
71
CCP
Base LLM=Mistral 7b
2024.03
66
CCP
Base LLM=Vicuna 13b
2024.03
66
Maximum Prob.
Base LLM=Jais 13b
2024.03
64
Token Entropy
Base LLM=Jais 13b
2024.03
63
Perplexity
Base LLM=Jais 13b
2024.03
61
P(True)
Base LLM=Vicuna 13b
2024.03
61
Maximum Prob.
Base LLM=Vicuna 13b
2024.03
60
Token Entropy
Base LLM=Mistral 7b
2024.03
60
Token Entropy
Base LLM=Vicuna 13b
2024.03
60
Maximum Prob.
Base LLM=Mistral 7b
2024.03
59
CCP
Base LLM=GPT-3.5-turbo
2024.03
58
Perplexity
Base LLM=Mistral 7b
2024.03
58
Perplexity
Base LLM=Vicuna 13b
2024.03
58
P(True)
Base LLM=Jais 13b
2024.03
55
Maximum Prob.
Base LLM=GPT-3.5-turbo
2024.03
54
Perplexity
Base LLM=GPT-3.5-turbo
2024.03
53
Token Entropy
Base LLM=GPT-3.5-turbo
2024.03
53
P(True)
Base LLM=Mistral 7b
2024.03
53
P(True)
Base LLM=GPT-3.5-turbo
2024.03
53
Feedback
Search any
task
Search any
task