Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Question Answering Retrieval on STaRK human-generated
Loading...
55.8
Hit@1
AF-Retriever
5.672
18.686
31.7
44.714
May 14, 2025
Hit@1
Hit@5
MRR
Updated 19d ago
Evaluation Results
Method
Method
Links
Hit@1
Hit@5
MRR
AF-Retriever
LLM Backbone=GPT OSS 1...
2025.05
55.8
66.4
60.7
4StepFocus
Data Version=10%-version
2025.05
52.9
61.5
50.3
KAR
Protocol=Zero-shot / O...
2025.05
52.6
63.9
57.6
AvaTaR
Protocol=Zero-shot / O...
2025.05
41.6
56.9
48.5
VSS + Reranker
Data Version=10%-version
2025.05
39.9
55.5
46.4
ReAct
Protocol=Zero-shot / O...
2025.05
31.6
48.4
40.3
Reflexion
Protocol=Zero-shot / O...
2025.05
31.5
45.5
39.8
VSS (Ada-002)
Embedding=Ada-002
2025.05
28.5
46.8
38.3
DPR
Protocol=Zero-shot / O...
2025.05
7.6
19.4
14.1
Feedback
Search any
task
Search any
task