List your company

Goal Prediction on EgoToM (test)

78.2True RateQwen2.5-VL-7B
5.71224.53143.3562.169Mar 25, 2026

Evaluation results

MethodLinks
Qwen2.5-VL-7BIntervention Setting=+αΔ2026.0378.245.935.4
Qwen2.5-VL-7BIntervention Setting=Baseline2026.0376.114.29.4
Gemini-2.5-FlashIntervention Setting=Baseline2026.0375.535.320.2
LLaVA-Next-Video-7BIntervention Setting=+αΔ2026.0327.399.927.2
LLaVA-Next-Video-7BIntervention Setting=+αΔ2026.0325.999.525.8
Qwen2.5-VL-7BIntervention Setting=+αΔ2026.032495.722.7
Gemini-2.5-FlashIntervention Setting=Baseline2026.0321.810021.8
Qwen2.5-VL-7BIntervention Setting=Baseline2026.0319.298.918.6
LLaVA-Next-Video-7BIntervention Setting=Baseline2026.0314.499.714.4
LLaVA-Next-Video-7BIntervention Setting=Baseline2026.038.51008.5