List your company

Correlation with human evaluation on SEEDA Fluency (S)

0.975Pearson Correlation (r)IMPARA
-0.65364-0.230820.1920.61482Feb 13, 2025

Evaluation results

MethodLinks
IMPARAaggregation_procedure=w/ TrueSkill2025.020.9750.952
SOMEaggregation_procedure=w/ TrueSkill2025.020.9710.925
GPT-4 S (fluency)2025.020.9520.916
GPT-4 S (meaning)2025.020.9520.925
SOMEaggregation_procedure=w/o TrueSkill2025.020.9310.916
IMPARAaggregation_procedure=w/o TrueSkill2025.020.8870.938
GPT-4 E (fluency)2025.020.7930.908
Scribendiaggregation_procedure=w/ TrueSkill2025.020.7450.859
Scribendiaggregation_procedure=w/o TrueSkill2025.020.6040.714
GREENaggregation_procedure=w/o TrueSkill2025.020.1850.569
GLEU+aggregation_procedure=w/o TrueSkill2025.020.1550.543
GLEU+aggregation_procedure=w/ TrueSkill2025.020.0170.393
GREENaggregation_procedure=w/ TrueSkill2025.02-0.2140.327
PTERRANTaggregation_procedure=w/ TrueSkill2025.02-0.3660.182
ERRANTaggregation_procedure=w/ TrueSkill2025.02-0.4630.095
PTERRANTaggregation_procedure=w/o TrueSkill2025.02-0.5460.077
ERRANTaggregation_procedure=w/o TrueSkill2025.02-0.591-0.156