AIMar 10

GenePlan: Evolving Better Generalized PDDL Plans using Large Language Models

arXiv:2603.09481v149.01 citationsh-index: 8
Predicted impact top 73% in AI · last 90 daysOriginality Incremental advance
AI Analysis

This work addresses the challenge of creating efficient and cost-effective generalized planners for AI planning domains, representing an incremental improvement by combining existing methods in a novel way.

The paper tackles the problem of generating generalized planners for classical planning tasks by introducing GenePlan, a framework that uses LLM-assisted evolutionary algorithms to evolve Python planners, achieving an average SAT score of 0.91, closely matching state-of-the-art planners at 0.93, and outperforming LLM-based baselines like chain-of-thought prompting at 0.64.

We present GenePlan (GENeralized Evolutionary Planner), a novel framework that leverages large language model (LLM) assisted evolutionary algorithms to generate domain-dependent generalized planners for classical planning tasks described in PDDL. By casting generalized planning as an optimization problem, GenePlan iteratively evolves interpretable Python planners that minimize plan length across diverse problem instances. In empirical evaluation across six existing benchmark domains and two new domains, GenePlan achieved an average SAT score of 0.91, closely matching the performance of the state-of-the-art planners (SAT score 0.93), and significantly outperforming other LLM-based baselines such as chain-of-thought (CoT) prompting (average SAT score 0.64). The generated planners solve new instances rapidly (average 0.49 seconds per task) and at low cost (average $1.82 per domain using GPT-4o).

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes