Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Opinion Alignment on smartvote 2023 Swiss national elections (test)
Loading...
70.73
Mean Macro-F1
SFT+GRPO
21.4132
34.2166
47.02
59.8234
Mar 1, 2026
Mean Macro-F1
Updated 3mo ago
Evaluation Results
Method
Method
Links
Mean Macro-F1
SFT+GRPO
Base model=Magistral 24B
2026.03
70.73
SFT
Base model=Magistral 24B
2026.03
67.63
SFT+GRPO
Base model=Llama 3.1 8B
2026.03
66.88
icl
Base model=Magistral 24B
2026.03
66.16
SFT+GRPO
Base model=Qwen3 8B
2026.03
65.11
SFT
Base model=Llama 3.1 8B
2026.03
63.44
SFT
Base model=Qwen3 8B
2026.03
61.08
GRPO
Base model=Qwen3 8B
2026.03
60.64
GRPO
Base model=Magistral 24B
2026.03
60.56
icl
Base model=Qwen3 8B
2026.03
60.48
icl
Base model=Llama 3.1 8B
2026.03
55.97
GRPO
Base model=Llama 3.1 8B
2026.03
55.14
random
Base model=Untrained b...
2026.03
50
ORPO
Base model=Llama 3.1 8B
2026.03
43.53
majority
Base model=Untrained b...
2026.03
37.43
ORPO
Base model=Qwen3 8B
2026.03
23.87
ORPO
Base model=Magistral 24B
2026.03
23.31
Feedback
Search any
task
Search any
task