Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Question Answering on PopQA (EM, FBERT)
Loading...
47.87
EM
Dynamic Collaboration Framework
12.874
21.9595
31.045
40.1305
Apr 20, 2026
EM
FBERT
Updated 1mo ago
Evaluation Results
Method
Method
Links
EM
FBERT
Dynamic Collaboration Framework
SLM=Qwen3-8B
2026.04
47.87
75.58
Dynamic Collaboration Framework
SLM=Llama-3.1-8B
2026.04
45.54
74.43
CoT
SLM=Qwen3-235B
2026.04
45.44
73.08
Dynamic Collaboration Framework
SLM=Qwen3-4B
2026.04
44.17
74.08
Dynamic Collaboration Framework
SLM=Llama-3.2-3B
2026.04
41.5
72.17
Direct Infer
SLM=Qwen3-235B
2026.04
38.38
67.64
CoT
SLM=Llama-3.1-8B
2026.04
25.59
61.45
CoT
SLM=Qwen3-8B
2026.04
20.69
58.95
CoT
SLM=Qwen3-4B
2026.04
16.04
57.28
CoT
SLM=Llama-3.2-3B
2026.04
14.22
52.65
Feedback
Search any
task
Search any
task