List your company

Multiple-choice Question Answering on HellaSwag (test)

80.1AccuracyBase
23.721638.358352.99567.6317Feb 24, 2026Mar 11, 2026Mar 26, 2026Apr 10, 2026Apr 25, 2026May 10, 2026May 26, 2026

Evaluation results

MethodLinks
BaseBackbone=QWEN2-7B2026.0580.1
PPT (Distribution)Backbone=QWEN2-7B2026.0579.4
PPT (Mean)Backbone=QWEN2-7B2026.0579.3
PPT (Distribution)Backbone=LLAMA3-8B2026.0566.8
BaseBackbone=LLAMA3-8B2026.0563.6
PPT (Mean)Backbone=LLAMA3-8B2026.0561.6
VerbalizedBackbone=LLAMA3-8B2026.0543.1
MDLM2026.0231.54
AR Transformer2026.0230.55
Duo++k=52026.0226.9
Duo2026.0226.46
Duo++k=22026.0226.26
Duo++k=32026.0225.89