List your company

Counterfactual Generation on TAXI1500

93.7SLFRGemma3-12B
7.48429.86752.2574.633May 12, 2026

Evaluation results

MethodLinks
Gemma3-12BLanguage=de, Protocol=TB-CF2026.0593.766.713.0125.78.387-7.387
Gemma3-4BLanguage=ar, Protocol=TB-CF2026.0591.933.317.7932.711.583-10.583
Gemma3-12BLanguage=sw, Protocol=TB-CF2026.0591.976.613.5447.67.483-6.483
Gemma3-4BLanguage=zh, Protocol=TB-CF2026.0589.231.514.6429.126.717-25.717
Gemma3-12BLanguage=vi, Protocol=TB-CF2026.0589.255.912.0824.78.459-7.459
Gemma3-12BLanguage=ru, Protocol=TB-CF2026.0588.360.415.7527.19.984-8.984
Gemma3-4BLanguage=de, Protocol=TB-CF2026.0587.432.411.5929.38.018-7.018
Gemma3-4BLanguage=sw, Protocol=TB-CF2026.0586.535.114.7850.37.537-6.537
Gemma3-4BLanguage=ru, Protocol=TB-CF2026.0586.528.812.7331.79.932-8.932
Gemma3-4BLanguage=vi, Protocol=TB-CF2026.0583.82710.4629.57.942-6.942
Gemma3-12BLanguage=zh, Protocol=TB-CF2026.0583.86417.425.726.798-25.798
Gemma3-12BLanguage=en, Protocol=DG-CF2026.0583.273.641.4843.70.7310.269
Gemma3-12BLanguage=ar, Protocol=TB-CF2026.0580.263.122.2229.811.946-10.946
Gemma3-12BLanguage=de, Protocol=DG-CF2026.0579.167.826.553.70.6860.314
MACROModel=Gemma3-12B, Language=avg2026.0579—23.860——
Gemma3-12BLanguage=ar, Protocol=DG-CF2026.0577.767.237.2553.70.6990.301
DG-CFModel=Gemma3-12B, Language=avg2026.0575—26.855——
Gemma3-12BLanguage=ru, Protocol=DG-CF2026.0574.964.320.49600.7430.257
Gemma3-12BLanguage=vi, Protocol=DG-CF2026.0574.46516.1851.70.7070.293
SFTModel=Gemma3-12B, Language=avg2026.0571—21.558——
Gemma3-12BLanguage=sw, Protocol=DG-CF2026.0569.457.817.4668.40.6530.347
Gemma3-12BLanguage=zh, Protocol=DG-CF2026.0568.260.527.954.40.850.15
Qwen3-4BLanguage=ar, Protocol=DG-CF2026.0568.134.286.1557.70.6150.385
Qwen3-4BLanguage=vi, Protocol=TB-CF2026.0567.628.819.1856.50.8140.186
Qwen3-8BLanguage=ar, Protocol=TB-CF2026.0566.733.341.3600.8080.192
Qwen3-8BLanguage=ar, Protocol=DG-CF2026.0565.838.750.2963.90.6220.378
Qwen3-4BLanguage=vi, Protocol=DG-CF2026.0565.135.625.67660.4850.515
Qwen3-4BLanguage=ar, Protocol=TB-CF2026.0564.927.955.9856.30.8390.161
MACROModel=Qwen3-8B, Language=avg2026.0564—24.970——
Qwen3-4BLanguage=ru, Protocol=TB-CF2026.0563.132.425.04620.8480.152
Qwen3-8BLanguage=zh, Protocol=DG-CF2026.0563.13638.9266.20.6890.311
Qwen3-4BLanguage=en, Protocol=DG-CF2026.0562.348.532.6461.40.480.52
Qwen3-4BLanguage=zh, Protocol=TB-CF2026.0562.230.635.72600.9270.073
MACROModel=Qwen3-4B, Language=avg2026.0562—28.970——
SFTModel=Qwen3-8B, Language=avg2026.0562—22.665——
MACROModel=Gemma3-4B, Language=en2026.0560—23.559——
MACROModel=Gemma3-4B, Language=avg2026.0560—23.568——
SFTModel=Qwen3-4B, Language=avg2026.0560—2866——
Qwen3-4BLanguage=zh, Protocol=DG-CF2026.0558.730.945.8467.40.5890.411
Qwen3-8BLanguage=zh, Protocol=TB-CF2026.0557.732.433.3763.90.9330.067
Qwen3-8BLanguage=vi, Protocol=DG-CF2026.0557.740.518.8361.30.6210.379
DG-CFModel=Qwen3-4B, Language=avg2026.0557—40.567——
Gemma3-4BLanguage=sw, Protocol=DG-CF2026.0556.935.725.59670.6470.353
SFTModel=Gemma3-4B, Language=en2026.0556—23.459——
SFTModel=Gemma3-4B, Language=avg2026.0556—2465——
Qwen3-8BLanguage=en, Protocol=DG-CF2026.0555.938.730.9568.20.4740.526
Gemma3-4BLanguage=de, Protocol=DG-CF2026.0555.740.533.9363.90.580.42
Gemma3-4BLanguage=ru, Protocol=DG-CF2026.0553.239.421.4965.60.650.35
Gemma3-4BLanguage=en, Protocol=DG-CF2026.0553.143.436.2450.70.670.33
DG-CFModel=Gemma3-4B, Language=en2026.0553—36.251——
DG-CFModel=Gemma3-4B, Language=avg2026.0553—3162——
Qwen3-4BLanguage=ru, Protocol=DG-CF2026.0552.931.527.7367.50.5260.474
Gemma3-4BLanguage=zh, Protocol=DG-CF2026.0552.340.539.4861.20.8070.193
Gemma3-4BLanguage=ar, Protocol=DG-CF2026.0552.234.241.8862.80.7160.284
Qwen3-8BLanguage=vi, Protocol=TB-CF2026.0551.430.616.3660.30.8490.151
Qwen3-4BLanguage=sw, Protocol=TB-CF2026.0550.511.714.955.31.765-0.765
Gemma3-4BLanguage=vi, Protocol=DG-CF2026.0550.437.718.4761.30.6310.369
DG-CFModel=Qwen3-8B, Language=avg2026.0550—28.967——
Qwen3-8BLanguage=de, Protocol=DG-CF2026.0549.534.225.8561.70.5620.438
Qwen3-4BLanguage=de, Protocol=DG-CF2026.0548.628.440.1766.20.4440.556
Qwen3-4BLanguage=de, Protocol=TB-CF2026.0548.634.237.858.90.7310.269
Qwen3-4BLanguage=sw, Protocol=DG-CF2026.0546.310.725.61840.3120.688
Qwen3-8BLanguage=ru, Protocol=DG-CF2026.05452718.7367.60.5660.434
Qwen3-8BLanguage=de, Protocol=TB-CF2026.0543.225.226.4664.60.7270.273
Qwen3-8BLanguage=ru, Protocol=TB-CF2026.0537.828.818.3367.10.8470.153
Qwen3-8BLanguage=sw, Protocol=TB-CF2026.05185.418.9611.241-0.241
Qwen3-8BLanguage=sw, Protocol=DG-CF2026.0510.80.918.9683.30.6930.307