List your company

Long-form generation on Long-form generation datasets 1D-SameTask - OOD

0.16PRRSATMD-MSP
-0.1104-0.04020.030.1002Apr 13, 2026

Evaluation results

MethodLinks
SATMD-MSPModel Backbone=Gemma-2-9B2026.040.16
HUQ-SATMDModel Backbone=Gemma-2-9B2026.040.12
Lookback-lensModel Backbone=Llama 3.1-8B2026.040.1
SAPLMA (middle)Model Backbone=Gemma-2-9B2026.040.1
Lookback-lensModel Backbone=Gemma-2-9B2026.040.1
SAPLMA (top)Model Backbone=Gemma-2-9B2026.040.08
SATMDModel Backbone=Gemma-2-9B2026.040.07
Uhead (1 layer)Model Backbone=Llama 3.1-8B2026.040.05
SAPLMA (middle)Model Backbone=Llama 3.1-8B2026.040.04
SAPLMA (top)Model Backbone=Llama 3.1-8B2026.040.04
Uhead (2 layers)Model Backbone=Llama 3.1-8B2026.040.02
SATRMD-MSPModel Backbone=Gemma-2-9B2026.040.02
SATRMDModel Backbone=Gemma-2-9B2026.040.01
HUQ-SATRMDModel Backbone=Gemma-2-9B2026.040.01
SATMD-MSPModel Backbone=Llama 3.1-8B2026.040
SATRMD-MSPModel Backbone=Llama 3.1-8B2026.04-0.01
MSPModel Backbone=Llama 3.1-8B2026.04-0.01
HUQ-SATMDModel Backbone=Llama 3.1-8B2026.04-0.03
SATMDModel Backbone=Llama 3.1-8B2026.04-0.07
HUQ-SATRMDModel Backbone=Llama 3.1-8B2026.04-0.07
Uhead (1 layer)Model Backbone=Gemma-2-9B2026.04-0.07
Uhead (2 layers)Model Backbone=Gemma-2-9B2026.04-0.07
SATRMDModel Backbone=Llama 3.1-8B2026.04-0.08
MSPModel Backbone=Gemma-2-9B2026.04-0.1