Vote-Tree-Planner: Optimizing Execution Order in LLM-based Task Planning Pipeline via Voting

2025-02-13

AI Summary

To address the inefficiency and poor executability caused by redundant LLM invocations in LLM-driven embodied robot task planning, this paper proposes a voting-weighted tree-based planning method. The core innovation lies in explicitly modeling path credibility as a dynamic priority, guiding high-confidence path traversal during tree search via lightweight voting aggregation and dynamic path pruning—thereby significantly reducing redundant LLM queries. Our approach integrates the Prog-Prompt framework with structured tree-based planning, preserving task accuracy without compromise. Evaluated on unseen datasets, it achieves a 12.7% average improvement in task success rate, a 9.4% increase in target-condition recall, and a 38.5% reduction in LLM query count. Moreover, the method demonstrates markedly enhanced robustness and execution stability.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Intelligent Robots: Motion and Path Planning
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Search Tool Learning with LLM: Teaching LLMs to invoke search and make use of retrieved information
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Integrating large language models (LLMs) into closed-loop robotic task planning has become increasingly popular within embodied artificial intelligence. Previous efforts mainly focused on leveraging the strong reasoning abilities of LLMs to enhance task planning performance while often overlooking task planning efficiency and executability due to repetitive queries to LLMs. This paper addresses the synergy between LLMs and task planning systems, aiming to minimize redundancy while enhancing planning effectiveness. Specifically, building upon Prog-Prompt and the high-level concept of Tree-Planner, we propose Vote-Tree-Planner. This sampling strategy utilizes votes to guide plan traversal during the decision-making process. Our approach is motivated by a straightforward observation: assigning weights to agents during decision-making enables the evaluation of critical paths before execution. With this simple vote-tree construction, our method further improves the success rate and reduces the number of queries to LLMs. The experimental results highlight that our Vote-Tree-Planner demonstrates greater stability and shows a higher average success rate and goal condition recall on the unseen dataset compared with previous baseline methods. These findings underscore the potential of the Vote-Tree-Planner to enhance planning accuracy, reliability, and efficiency in LLM-based planning systems.

Problem

• Optimize task planning execution order
• Reduce LLM query redundancy
• Enhance planning effectiveness and efficiency

Innovation

• Vote-Tree-Planner strategy
• Minimizes LLM query redundancy
• Enhances planning success rate