Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Knowledge Retrieval on STARK-PRIME official (test)
Loading...
18.44
Hit@1
AVATAR
3.9008
7.6754
11.45
15.2246
Jun 17, 2024
Hit@1
Hit@5
R@20
MRR
Updated 3d ago
Evaluation Results
Method
Method
Links
Hit@1
Hit@5
R@20
MRR
AVATAR
2024.06
18.44
36.73
39.31
26.73
ReAct
2024.06
15.28
31.95
33.63
22.76
multi-ada-002
2024.06
15.1
33.56
38.05
23.49
Reflexion
2024.06
14.28
34.99
38.52
24.82
ada-002
2024.06
12.63
31.49
36
21.41
QAGNN
2024.06
8.85
21.35
29.63
14.73
AVATAR-C
Ablation=Removes the c...
2024.06
8.82
23.82
30.32
16.2
DPR
2024.06
4.46
21.85
30.13
12.38
Feedback
Search any
task
Search any
task