Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multi-Hop Question Answering on MuSiQuE (EM, FBERT)
Loading...
23.62
EM
CoT
6.4496
10.9073
15.365
19.8227
Apr 20, 2026
EM
FBERT Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
EM
FBERT Score
CoT
SLM=Qwen3-235B
2026.04
23.62
66.23
Dynamic Collaboration Framework
SLM=Qwen3-8B
2026.04
23.31
67.34
Dynamic Collaboration Framework
SLM=Qwen3-4B
2026.04
18.61
65
Direct Infer
SLM=Qwen3-235B
2026.04
18.15
62.46
Dynamic Collaboration Framework
SLM=Llama-3.1-8B
2026.04
17.79
63.7
Dynamic Collaboration Framework
SLM=Llama-3.2-3B
2026.04
14.47
60.61
CoT
SLM=Qwen3-8B
2026.04
12.93
58.38
CoT
SLM=Qwen3-4B
2026.04
10.12
57.36
CoT
SLM=Llama-3.1-8B
2026.04
9.41
50.88
CoT
SLM=Llama-3.2-3B
2026.04
7.11
50.02
Feedback
Search any
task
Search any
task