Decompose, Plan in Parallel, and Merge: A Novel Paradigm for Large Language Models based Planning with Multiple Constraints

TL;DR

This paper introduces DPPM, a parallel planning framework for LLMs that decomposes complex tasks, plans in parallel for subtasks, and merges them, effectively handling constraints and reducing errors in planning.

Contribution

The paper presents a novel parallel planning paradigm for LLMs that improves handling of constraints and reduces cascading errors in planning tasks.

Findings

DPPM outperforms existing methods in travel planning tasks.

Incorporates verification and refinement for error correction.

Effectively manages complex constraints in planning.

Abstract

Despite significant advances in Large Language Models (LLMs), planning tasks still present challenges for LLM-based agents. Existing planning methods face two key limitations: heavy constraints and cascading errors. To address these limitations, we propose a novel parallel planning paradigm, which Decomposes, Plans for subtasks in Parallel, and Merges subplans into a final plan (DPPM). Specifically, DPPM decomposes the complex task based on constraints into subtasks, generates the subplan for each subtask in parallel, and merges them into a global plan. In addition, our approach incorporates a verification and refinement module, enabling error correction and conflict resolution. Experimental results demonstrate that DPPM significantly outperforms existing methods in travel planning tasks.