ResearchBenchmarksTask Planning on RestBench TMDB v1 (test)Follow82.56n-F1Direct28.802442.758756.71570.6713May 29, 2024Evaluation ResultsMethodMethodLinksn-F1L-F1Output Tokens (x10^3)DirectLLM=GPT 4-turboLLM=GPT 4-turbo2024.0582.5656.671,750SGCLLM=GPT 4-turboLLM=GPT 4-turbo2024.0582.2756.371,620SGCLLM=Baichuan2 13BLLM=Baichuan2 13B2024.0532.979.111,840DirectLLM=Baichuan2 13BLLM=Baichuan2 13B2024.0530.879.921,960