Programming over Thinking: Efficient and Robust Multi-Constraint Planning

TL;DR

This paper introduces SCOPE, a framework that separates reasoning from code execution in multi-constraint planning, leading to more consistent, efficient, and generalizable solutions with state-of-the-art performance.

Contribution

SCOPE is a novel framework that disentangles reasoning from execution, enabling reusable, deterministic solver functions for multi-constraint planning.

Findings

Achieves 93.1% success on TravelPlanner, a 61.6% improvement over baseline.

Reduces inference cost by 1.4x and time by approximately 4.67x.

Outperforms existing approaches in multi-constraint planning tasks.

Abstract

Multi-constraint planning involves identifying, evaluating, and refining candidate plans while satisfying multiple, potentially conflicting constraints. Existing large language model (LLM) approaches face fundamental limitations in this domain. Pure reasoning paradigms, which rely on long natural language chains, are prone to inconsistency, error accumulation, and prohibitive cost as constraints compound. Conversely, LLMs combined with coding- or solver-based strategies lack flexibility: they often generate problem-specific code from scratch or depend on fixed solvers, failing to capture generalizable logic across diverse problems. To address these challenges, we introduce the Scalable COde Planning Engine (SCOPE), a framework that disentangles query-specific reasoning from generic code execution. By separating reasoning from execution, SCOPE produces solver functions that are consistent, deterministic, and reusable across queries while requiring only minimal changes to input parameters. SCOPE achieves state-of-the-art performance while lowering cost and latency. For example, with GPT-4o, it reaches 93.1% success on TravelPlanner, a 61.6% gain over the best baseline (CoT) while cutting inference cost by 1.4x and time by ~4.67x. Code is available at https://github.com/DerrickGXD/SCOPE.