Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Question Answering on HotpotQA (Accuracy and Latency)
Loading...
32
Accuracy
ToT
20.56
23.53
26.5
29.47
Jun 13, 2024
Accuracy
Latency (s)
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
Latency (s)
ToT
Backbone=LLAMA2-13B
2024.06
32
1,271
TS-SFT
Backbone=LLAMA2-13B
2024.06
30.3
65.5
CPO
Backbone=LLAMA2-13B
2024.06
30.3
63.8
ToT
Backbone=Mistral-7B
2024.06
30
4,698
CPO
Backbone=Mistral-7B
2024.06
29.4
56.9
CoT
Backbone=LLAMA2-13B
2024.06
29
65.2
TS-SFT
Backbone=Mistral-7B
2024.06
28.6
56.2
CoT
Backbone=Mistral-7B
2024.06
28
58.4
CPO
Backbone=LLAMA2-7B
2024.06
24
41.1
ToT
Backbone=LLAMA2-7B
2024.06
23
1,100.7
TS-SFT
Backbone=LLAMA2-7B
2024.06
22.7
44.8
CoT
Backbone=LLAMA2-7B
2024.06
21
45.5
Feedback
Search any
task
Search any
task