ResearchBenchmarksRobot Task Planning on Overall Combined TasksFollow99Token Cost (k)LLMTAMP73.96242.98412581.02May 23, 2025Evaluation ResultsMethodMethodLinksToken Cost (k)Success RateLLMTAMPLLM Backbone=GPT-4oLLM Backbone=GPT-4o2025.0599—PDDLLMExecution=PDDL solverExecution=PDDL solver2025.05415—O1-TAMPLLM Backbone=OpenAI o1LLM Backbone=OpenAI o12025.05666—R1-TAMPLLM Backbone=DeepSeek R1LLM Backbone=DeepSeek R12025.05725—