List your company

Step Completion on CoSPlan Blocks-World-E

54.3AccuracyGPT-4o
16.65226.42636.245.974Dec 11, 2025

Evaluation results

MethodLinks
GPT-4oReasoning Strategy=Scene Graph2025.1254.3
GPT-4oReasoning Strategy=Chain-of-Thought2025.1249.7
CoG-VLMReasoning Strategy=Scene Graph2025.1226.7
CoG-VLMReasoning Strategy=Vanilla2025.1225.5
CoG-VLMReasoning Strategy=Chain-of-Thought2025.1225.2
Janus-pro-7BReasoning Strategy=Scene Graph2025.1225.1
Janus-pro-7BReasoning Strategy=Vanilla2025.1224.2
Janus-pro-7BReasoning Strategy=Chain-of-Thought2025.1223.1
Intern-VLMReasoning Strategy=Chain-of-Thought2025.1221.2
Random2025.1220
Intern-VLMReasoning Strategy=Scene Graph2025.1218.9
Qwen2 VL-8BReasoning Strategy=Scene Graph2025.1218.8
Qwen2 VL-8BReasoning Strategy=Chain-of-Thought2025.1218.6
Intern-VLMReasoning Strategy=Vanilla2025.1218.3
Qwen2 VL-8BReasoning Strategy=Vanilla2025.1218.1