🤖 AI Summary
Addressing the challenges of enhancing complex logical reasoning capabilities and the scarcity of high-quality training data, this paper proposes LogicPro—a novel methodology for automated, verifiable, and scalable reasoning-chain construction. LogicPro pioneers a program-guided reasoning synthesis paradigm, leveraging intermediate variables from LeetCode-style algorithmic problems and their Python implementations as explicit reasoning anchors. It integrates program parsing, test-case-driven synthesis, and structured code-to-text mapping to generate 540K high-fidelity reasoning samples from only 2,360 problems. Evaluated on diverse benchmarks—including BBH, LogicBench, DROP, AR-LSAT, and GSM8K—LogicPro consistently improves the logical reasoning performance of mainstream large language models. Results demonstrate that program-instructed learning effectively enhances both the fidelity and generalizability of complex reasoning modeling, offering a principled, data-efficient pathway toward robust logical reasoning capability acquisition.
📝 Abstract
In this paper, we propose a new data synthesis method called extbf{LogicPro}, which leverages LeetCode-style algorithm underline{Pro}blems and their corresponding underline{Pro}gram solutions to synthesize Complex underline{Logic}al Reasoning data in text format. First, we synthesize complex reasoning problems through source algorithm problems and test cases. Then, standard answers and intermediate variable outputs are obtained for each problem based on standard python solutions and test cases. Finally, with the guidance of code intermediate variables, we synthesize the text reasoning process for each reasoning problems. Through this method, we can synthesize data that is difficult, scalable, effective, and comes with golden standard answers and high-quality reasoning processes. As a result, with our 540K synthesized dataset constructed solely from 2,360 algorithm problems, our approach Code and data are publicly available at https://github.com/jiangjin1999/LogicPro achieves significant improvements in multiple models for the datasets extit{BBH$^{27}$}, extit{LogicBench}, extit{DROP}, extit{AR-LSAT}, and extit{GSM8K}, etc. outperforming a wide range of existing reasoning datasets.