List your company

Language-driven scene representation on ALFRED Object Shift [OS]

83.92F1 ScoreFalcon-7B
67.5471.792576.04580.2975May 7, 2026

Evaluation results

MethodLinks
Falcon-7BApproach=CLM2026.0583.9266.382.34
Mistral-7BApproach=CLM2026.0582.7263.462.75
LLaMA 3.1-8BApproach=CLM2026.0580.4458.864.27
Vicuna-7BApproach=IM2026.0580.0351.674.4
AlpacaApproach=IM2026.0577.2743.984.64
Flan-T5Approach=Seq2Seq2026.0576.6747.433.48
T5Approach=Seq2Seq2026.0568.1736.994.51