dorsal/arxiv
View SchemaProgramming over Thinking: Efficient and Robust Multi-Constraint Planning
| Authors | Derrick Goh Xin Deik, Quanyu Long, Zhengyuan Liu, Nancy F. Chen, Wenya Wang |
|---|---|
| Categories | |
| ArXiv ID | 2601.09097vv2 |
| URL | https://arxiv.org/abs/2601.09097 |
| License | http://creativecommons.org/licenses/by-nc-sa/4.0/ |
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.
{
"annotation_id": "5474ae18-a367-4f83-8403-6a63db6a8870",
"date_created": "2026-02-17T05:53:19.842000Z",
"date_modified": "2026-02-17T05:53:19.842000Z",
"file_hash": "a9b503a7b3bfc965f31f754ec384f7a355e648c287fefa103763ccd62a43a692",
"private": false,
"record": {
"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.",
"arxiv_id": "2601.09097",
"authors": [
"Derrick Goh Xin Deik",
"Quanyu Long",
"Zhengyuan Liu",
"Nancy F. Chen",
"Wenya Wang"
],
"categories": [
"cs.AI"
],
"license": "http://creativecommons.org/licenses/by-nc-sa/4.0/",
"title": "Programming over Thinking: Efficient and Robust Multi-Constraint Planning",
"url": "https://arxiv.org/abs/2601.09097",
"version": "v2"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "490f50c4-8100-45e6-ac15-48adbec01f1f",
"id": "arXiv Dataset",
"type": "Model",
"variant": "snapshot-2026-01-17",
"version": "0.1.0"
},
"user_id": 1000002
}