Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Question Answering Feedback Generation on QA-FEEDBACK (test)
Loading...
0.513
Rs1 (Relevance)
SFT
0.48076
0.48913
0.4975
0.50587
Jun 2, 2023
Rs1 (Relevance)
Rs2 (Factuality)
Rs3 (Completeness)
ROUGE Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Rs1 (Relevance)
Rs2 (Factuality)
Rs3 (Completeness)
ROUGE Score
SFT
training=Supervised Fi...
2023.06
0.513
0.749
-0.053
48.96
FINE-GRAINED RLHF
base_model=SFT, traini...
2023.06
0.513
0.816
0.139
49.93
SFT-Full
training=Supervised Fi...
2023.06
0.508
0.756
0.044
49.63
Pref. RLHF
base_model=SFT, traini...
2023.06
0.482
0.781
0.101
49.84
Feedback
Search any
task
Search any
task