List your company

Embodied Task Planning on ALFWorld standard evaluation set (134 tasks)

97Pass@1 AccuracyGiG
59.5669.287988.72Jan 29, 2026

Evaluation results

MethodLinks
GiGModel=Qwen32026.0197
GiGModel=DeepSeek2026.0197
GiGModel=Gemini2026.0191
ReCAPModel=Qwen32026.0189
ReCAPModel=Gemini2026.0186
ReCAPModel=DeepSeek2026.0182
ReActModel=Qwen32026.0161