Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Question Answering on Hotpot QA (val)
Loading...
24.6
EM
Llama2-7b CBQA w. Mixed Training
16.384
18.517
20.65
22.783
Mar 12, 2024
EM
F1
Rec-BLEU
Updated 4d ago
Evaluation Results
Method
Method
Links
EM
F1
Rec-BLEU
Llama2-7b CBQA w. Mixed Training
Model=Llama2-7b, Setup...
2024.03
24.6
34.6
-
Llama2-7b Closed-book QA
Model=Llama2-7b, Setup...
2024.03
24.1
33.8
-
Llama2-7b + Recitation
Model=Llama2-7b, Setup...
2024.03
22.7
31.4
22.2
Llama2-7b CBQA w. Mixed Training + Recitation
Model=Llama2-7b, Setup...
2024.03
22.4
30.7
23.8
Qwen1.5-4b CBQA w. Mixed Training
Model=Qwen1.5-4b, Setu...
2024.03
19.3
28.3
-
Qwen1.5-4b Closed-book QA
Model=Qwen1.5-4b, Setu...
2024.03
18.6
27
-
Qwen1.5-4b CBQA w. Mixed Training + Recitation
Model=Qwen1.5-4b, Setu...
2024.03
18.4
25.1
32.7
Qwen1.5-4b + Recitation
Model=Qwen1.5-4b, Setu...
2024.03
16.7
24.5
17.4
Feedback
Search any
task
Search any
task