List your company

Opinion Alignment on Wahl-O-Mat (WoM) March 2025 (test)

53.21Mean Macro-F1SFT+GRPO
23.590831.280438.9746.6596Mar 1, 2026

Evaluation results

MethodLinks
SFT+GRPOBase model=Magistral 24B2026.0353.21
SFT+GRPOBase model=Llama 3.1 8B2026.0352.53
SFTBase model=Magistral 24B2026.0351.86
GRPOBase model=Magistral 24B2026.0351
SFT+GRPOBase model=Qwen3 8B2026.0349.38
SFTBase model=Llama 3.1 8B2026.0348.95
ORPOBase model=Llama 3.1 8B2026.0343.29
SFTBase model=Qwen3 8B2026.0342.91
GRPOBase model=Llama 3.1 8B2026.0337.29
randomBase model=Untrained baselines2026.0333.33
GRPOBase model=Qwen3 8B2026.0331.42
iclBase model=Llama 3.1 8B2026.0328.17
majorityBase model=Untrained baselines2026.0327.44
iclBase model=Qwen3 8B2026.0326.19
iclBase model=Magistral 24B2026.0326.19
ORPOBase model=Qwen3 8B2026.0325.25
ORPOBase model=Magistral 24B2026.0324.73