List your company

Short-form generation on Short-form generation datasets LOO - near OOD

0.64PRRMSP
-0.07760.10870.2950.4813Apr 13, 2026

Evaluation results

MethodLinks
MSPModel Backbone=Gemma-2-9B2026.040.64
HUQ-SATRMDModel Backbone=Gemma-2-9B2026.040.62
HUQ-SATMDModel Backbone=Llama 3.1-8B2026.040.59
MSPModel Backbone=Llama 3.1-8B2026.040.57
HUQ-SATRMDModel Backbone=Llama 3.1-8B2026.040.55
HUQ-SATMDModel Backbone=Gemma-2-9B2026.040.52
SAPLMA (middle)Model Backbone=Llama 3.1-8B2026.040.46
SATMDModel Backbone=Llama 3.1-8B2026.040.41
SATRMD-MSPModel Backbone=Gemma-2-9B2026.040.4
SAPLMA (middle)Model Backbone=Gemma-2-9B2026.040.35
SAPLMA (top)Model Backbone=Gemma-2-9B2026.040.34
Uhead (1 layer)Model Backbone=Gemma-2-9B2026.040.34
Uhead (2 layers)Model Backbone=Gemma-2-9B2026.040.33
SATRMDModel Backbone=Gemma-2-9B2026.040.33
SATRMD-MSPModel Backbone=Llama 3.1-8B2026.040.3
SATMD-MSPModel Backbone=Llama 3.1-8B2026.040.28
SAPLMA (top)Model Backbone=Llama 3.1-8B2026.040.26
Uhead (1 layer)Model Backbone=Llama 3.1-8B2026.040.17
SATRMDModel Backbone=Llama 3.1-8B2026.040.17
SATMDModel Backbone=Gemma-2-9B2026.040.16
Uhead (2 layers)Model Backbone=Llama 3.1-8B2026.040.15
SATMD-MSPModel Backbone=Gemma-2-9B2026.040.13
Lookback-lensModel Backbone=Llama 3.1-8B2026.040
Lookback-lensModel Backbone=Gemma-2-9B2026.04-0.05