Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Hop Question Answering on HotpotQA (test val)
Loading...
47.1
Exact Match (EM)
DeepControl
13.612
22.306
31
39.694
Feb 2, 2026
Exact Match (EM)
Average EM
Updated 1mo ago
Evaluation Results
Method
Method
Links
Exact Match (EM)
Average EM
DeepControl
Backbone Model=Qwen2.5-7B
2026.02
47.1
47.9
DeepControl
Backbone Model=Qwen2.5-3B
2026.02
40.2
41.1
Direct Inference
Backbone Model=Qwen2.5-7B
2026.02
18.3
18.1
Direct Inference
Backbone Model=Qwen2.5-3B
2026.02
14.9
13.4
Feedback
Search any
task
Search any
task