ResearchBenchmarksFlip 4 on CONSTRUCTIONFlip 4 on CONSTRUCTIONFollow100Success RateGT61.5271.5181.591.49Dec 3, 2023Evaluation resultsMethodMethodLinksSuccess RateGTMode=Planning with gro...Mode=Planning with ground truth models2023.12100RNDMode=Zero-shotMode=Zero-shot2023.1282CEE-USMode=Zero-shotMode=Zero-shot2023.1273DisMode=Zero-shotMode=Zero-shot2023.1269RaIRMode=Zero-shotMode=Zero-shot2023.1265RaIR + CEE-USMode=Zero-shotMode=Zero-shot2023.1263