TIC: Translate-Infer-Compile for accurate "text to plan" using LLMs and Logical Representations
This addresses the challenge of bridging natural language and automated planning for users needing reliable text-to-plan systems, representing an incremental improvement over direct LLM approaches.
The paper tackles the problem of generating accurate plans from natural language requests by introducing TIC, a method that uses LLMs to produce an intermediate logical representation, then applies logical inference and compilation to PDDL, achieving high accuracy across seven domains.
We study the problem of generating plans for given natural language planning task requests. On one hand, LLMs excel at natural language processing but do not perform well on planning. On the other hand, classical planning tools excel at planning tasks but require input in a structured language such as the Planning Domain Definition Language (PDDL). We leverage the strengths of both the techniques by using an LLM for generating the PDDL representation (task PDDL) of planning task requests followed by using a classical planner for computing a plan. Unlike previous approaches that use LLMs for generating task PDDLs directly, our approach comprises of (a) translate: using an LLM only for generating a logically interpretable intermediate representation of natural language task description, (b) infer: deriving additional logically dependent information from the intermediate representation using a logic reasoner (currently, Answer Set Programming solver), and (c) compile: generating the target task PDDL from the base and inferred information. We observe that using an LLM to only output the intermediate representation significantly reduces LLM errors. Consequently, TIC approach achieves, for at least one LLM, high accuracy on task PDDL generation for all seven domains of our evaluation dataset.