List your company
ResearchFormerly WizWand

Predicting Language Model Output on IMDB

0.0089Brier ScoreLAMP
-0.0195840.1726830.364950.557217May 17, 2025
Updated 6mo ago

Evaluation results

MethodLinks
LAMPBase Language Model=Mistral Large2025.050.0089
SHAPBase Language Model=Mistral Large2025.050.0103
SHAPBase Language Model=Claude 3.5 Haiku2025.050.0194
Mean baselineBase Language Model=Mistral Large2025.050.021
SHAPBase Language Model=GPT-4.1-mini2025.050.0262
LAMPBase Language Model=Gemini 2.5 Flash2025.050.0276
SHAPBase Language Model=Gemini 2.5 Flash2025.050.0292
LAMPBase Language Model=GPT-4.1-mini2025.050.0303
LAMPBase Language Model=Claude 3.5 Haiku2025.050.0428
ANCHORSBase Language Model=GPT-4.1-mini2025.050.0576
Mean baselineBase Language Model=Claude 3.5 Haiku2025.050.072
Uniform baselineBase Language Model=Claude 3.5 Haiku2025.050.1161
Mean baselineBase Language Model=GPT-4.1-mini2025.050.1201
Uniform baselineBase Language Model=Mistral Large2025.050.1673
Uniform baselineBase Language Model=GPT-4.1-mini2025.050.1732
ANCHORSBase Language Model=Mistral Large2025.050.1777
ANCHORSBase Language Model=Claude 3.5 Haiku2025.050.1869
Uniform baselineBase Language Model=Gemini 2.5 Flash2025.050.2281
Mean baselineBase Language Model=Gemini 2.5 Flash2025.050.3016
ANCHORSBase Language Model=Gemini 2.5 Flash2025.050.3225
LIMEBase Language Model=GPT-4.1-mini2025.050.3609
Random baselineBase Language Model=Claude 3.5 Haiku2025.050.3776
Random baselineBase Language Model=Mistral Large2025.050.4005
Random baselineBase Language Model=GPT-4.1-mini2025.050.4015
Random baselineBase Language Model=Gemini 2.5 Flash2025.050.4354
LIMEBase Language Model=Claude 3.5 Haiku2025.050.572
LIMEBase Language Model=Mistral Large2025.050.6165
LIMEBase Language Model=Gemini 2.5 Flash2025.050.721