{"ID":2890391,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2507.19060","arxiv_id":"2507.19060","title":"PurpCode: Reasoning for Safer Code Generation","abstract":"We introduce PurpCode, the first post-training recipe for training safe code reasoning models towards generating secure code and defending against malicious cyberactivities. PurpCode trains a reasoning model in two stages: (i) Rule Learning, which explicitly teaches the model to reference cybersafety rules to generate vulnerability-free code and to avoid facilitating malicious cyberactivities; and (ii) Reinforcement Learning, which optimizes model safety and preserves model utility through diverse, multi-objective reward mechanisms. To empower the training pipelines with comprehensive cybersafety data, we conduct internal red-teaming to synthesize comprehensive and high-coverage prompts based on real-world tasks for inducing unsafe cyberactivities in the model. Based on PurpCode, we develop a reasoning-based coding model, namely PurpCode-32B, which demonstrates state-of-the-art cybersafety, outperforming various frontier models. Meanwhile, our alignment method decreases the model overrefusal rates in both general and cybersafety-specific scenarios, while preserving model utility in both code generation and common security knowledge.","short_abstract":"We introduce PurpCode, the first post-training recipe for training safe code reasoning models towards generating secure code and defending against malicious cyberactivities. PurpCode trains a reasoning model in two stages: (i) Rule Learning, which explicitly teaches the model to reference cybersafety rules to generate...","url_abs":"https://arxiv.org/abs/2507.19060","url_pdf":"https://arxiv.org/pdf/2507.19060v4","authors":"[\"Jiawei Liu\",\"Nirav Diwan\",\"Zhe Wang\",\"Haoyu Zhai\",\"Xiaona Zhou\",\"Kiet A. Nguyen\",\"Tianjiao Yu\",\"Muntasir Wahed\",\"Yinlin Deng\",\"Hadjer Benkraouda\",\"Yuxiang Wei\",\"Lingming Zhang\",\"Ismini Lourentzou\",\"Gang Wang\"]","published":"2025-07-25T08:23:00Z","proceeding":"cs.CR","tasks":"[\"cs.CR\",\"cs.CL\",\"cs.LG\",\"cs.SE\"]","methods":"[\"Reinforcement Learning\"]","has_code":false}