Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Multi-hop retrieval question answering on HotPotQA (test)
Loading...
45.6
Answer Accuracy
multihop
20.848
27.274
33.7
40.126
Oct 5, 2023
Answer Accuracy
Passage Retrieval Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Answer Accuracy
Passage Retrieval Score
multihop
Model=GPT-3.5, Compile...
2023.10
45.6
-
multihop
Model=Llama2-13b-chat,...
2023.10
41
-
multihop
Model=GPT-3.5, Compile...
2023.10
39.6
43.8
multihop
Model=Llama2-13b-chat,...
2023.10
36.4
43.5
COT_RAG
Model=Llama2-13b-chat,...
2023.10
32.9
34.4
vanilla
Model=GPT-3.5, Compile...
2023.10
31.5
-
multihop
Model=Llama2-13b-chat,...
2023.10
31.3
30.8
multihop
Model=GPT-3.5, Compile...
2023.10
31.2
40.8
COT_RAG
Model=GPT-3.5, Compile...
2023.10
29.8
34.4
COT_RAG
Model=Llama2-13b-chat,...
2023.10
28
34.4
vanilla
Model=Llama2-13b-chat,...
2023.10
21.8
-
Feedback
Search any
task
Search any
task