List your company

Short-form generation on Short-form generation datasets 1D-DiffTask most OOD

0.64PRRMSP
-0.4-0.130.140.41Apr 13, 2026

Evaluation results

MethodLinks
MSPModel Backbone=Gemma-2-9B2026.040.64
MSPModel Backbone=Llama 3.1-8B2026.040.57
SATRMD-MSPModel Backbone=Gemma-2-9B2026.040.52
SATRMDModel Backbone=Gemma-2-9B2026.040.34
SAPLMA (middle)Model Backbone=Llama 3.1-8B2026.040.32
SATMD-MSPModel Backbone=Gemma-2-9B2026.040.27
HUQ-SATMDModel Backbone=Llama 3.1-8B2026.040.24
HUQ-SATRMDModel Backbone=Llama 3.1-8B2026.040.22
SATRMD-MSPModel Backbone=Llama 3.1-8B2026.040.21
SAPLMA (middle)Model Backbone=Gemma-2-9B2026.040.21
Lookback-lensModel Backbone=Llama 3.1-8B2026.040.19
SAPLMA (top)Model Backbone=Llama 3.1-8B2026.040.18
HUQ-SATMDModel Backbone=Gemma-2-9B2026.040.17
HUQ-SATRMDModel Backbone=Gemma-2-9B2026.040.17
Uhead (1 layer)Model Backbone=Gemma-2-9B2026.040.16
Uhead (2 layers)Model Backbone=Gemma-2-9B2026.040.16
Lookback-lensModel Backbone=Gemma-2-9B2026.040.16
SATMDModel Backbone=Llama 3.1-8B2026.040.15
SATRMDModel Backbone=Llama 3.1-8B2026.040.13
Uhead (1 layer)Model Backbone=Llama 3.1-8B2026.040.08
SATMD-MSPModel Backbone=Llama 3.1-8B2026.040.08
SAPLMA (top)Model Backbone=Gemma-2-9B2026.040.08
SATMDModel Backbone=Gemma-2-9B2026.040.02
Uhead (2 layers)Model Backbone=Llama 3.1-8B2026.04-0.36