Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
General Question Answering on TriviaQA (EM, FBERT)
Loading...
66.3
EM
CoT
32.968
41.6215
50.275
58.9285
Apr 20, 2026
EM
FBERT Score
Updated 1mo ago
Evaluation Results
Method
Method
Links
EM
FBERT Score
CoT
SLM=Qwen3-235B
2026.04
66.3
85.95
Dynamic Collaboration Framework
SLM=Qwen3-8B
2026.04
64.64
85.32
Dynamic Collaboration Framework
SLM=Llama-3.1-8B
2026.04
62.25
84.06
Dynamic Collaboration Framework
SLM=Qwen3-4B
2026.04
61.82
83.95
Direct Infer
SLM=Qwen3-235B
2026.04
61.72
83.72
Dynamic Collaboration Framework
SLM=Llama-3.2-3B
2026.04
57.5
82.01
CoT
SLM=Qwen3-8B
2026.04
47.83
76.19
CoT
SLM=Llama-3.1-8B
2026.04
46
73.44
CoT
SLM=Qwen3-4B
2026.04
37.57
71.21
CoT
SLM=Llama-3.2-3B
2026.04
34.25
67.23
Feedback
Search any
task
Search any
task