List your company

Error Detection on CoSPlan Blocks-World-E

44.5AccuracyCoG-VLM
25.36430.33235.340.268Dec 11, 2025

Evaluation results

MethodLinks
CoG-VLMReasoning Strategy=Scene Graph2025.1244.5
CoG-VLMReasoning Strategy=Chain-of-Thought2025.1243.1
GPT-4oReasoning Strategy=Scene Graph2025.1242.1
CoG-VLMReasoning Strategy=Vanilla2025.1241.3
Intern-VLMReasoning Strategy=Chain-of-Thought2025.1237.9
Intern-VLMReasoning Strategy=Scene Graph2025.1237.3
Intern-VLMReasoning Strategy=Vanilla2025.1236.5
Qwen2 VL-8BReasoning Strategy=Scene Graph2025.1235.2
GPT-4oReasoning Strategy=Chain-of-Thought2025.1235.1
Qwen2 VL-8BReasoning Strategy=Vanilla2025.1232.3
Janus-pro-7BReasoning Strategy=Chain-of-Thought2025.1231
Qwen2 VL-8BReasoning Strategy=Chain-of-Thought2025.1230.6
Janus-pro-7BReasoning Strategy=Vanilla2025.1229.3
Janus-pro-7BReasoning Strategy=Scene Graph2025.1227.6
Random2025.1226.1