# PurpCode: Reasoning for Safer Code Generation

arXiv: https://arxiv.org/abs/2507.19060 · HTML: https://mlsanity.com/paper/2507.19060/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Placed in Reasoning for agents, planning and code, one of 17 directions in LLM reasoning.
- Builds on 60 of its 137 references that are in the map.
- First posted on arXiv 25 Jul 2025.

## Placed in

- LLM reasoning / [Reasoning for agents, planning and code](https://mlsanity.com/lr/reasoning-for-agents-planning-and-code/)

## Builds on (most cited first, 30 of 60)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning](https://mlsanity.com/paper/2501.12948/) (2025)
- [Qwen2.5 Technical Report](https://mlsanity.com/paper/2412.15115/) (2024)
- [DAPO: An Open-Source LLM Reinforcement Learning System at Scale](https://mlsanity.com/paper/2503.14476/) (2025)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [HarmBench: A Standardized Evaluation Framework for Automated Red Teaming and Robust Refusal](https://mlsanity.com/paper/2402.04249/) (2024)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [Qwen2.5-Coder Technical Report](https://mlsanity.com/paper/2409.12186/) (2024)
- [CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation](https://mlsanity.com/paper/2109.00859/) (2021)
- [CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis](https://mlsanity.com/paper/2203.13474/) (2022)
- [StarCoder: may the source be with you!](https://mlsanity.com/paper/2305.06161/) (2023)
- [AutoDAN: Generating Stealthy Jailbreak Prompts on Aligned Large Language Models](https://mlsanity.com/paper/2310.04451/) (2023)
- [WizardCoder: Empowering Code Large Language Models with Evol-Instruct](https://mlsanity.com/paper/2306.08568/) (2023)
- [Deduplicating Training Data Makes Language Models Better](https://mlsanity.com/paper/2107.06499/) (2021)
- [SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training](https://mlsanity.com/paper/2501.17161/) (2025)
- [BigCodeBench: Benchmarking Code Generation with Diverse Function Calls and Complex Instructions](https://mlsanity.com/paper/2406.15877/) (2024)
- [JailbreakBench: An Open Robustness Benchmark for Jailbreaking Large Language Models](https://mlsanity.com/paper/2404.01318/) (2024)
- [XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models](https://mlsanity.com/paper/2308.01263/) (2023)
- [Process Reinforcement through Implicit Rewards](https://mlsanity.com/paper/2502.01456/) (2025)
- [StarCoder 2 and The Stack v2: The Next Generation](https://mlsanity.com/paper/2402.19173/) (2024)
- [WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs](https://mlsanity.com/paper/2406.18495/) (2024)
- ["Do Anything Now": Characterizing and Evaluating In-The-Wild Jailbreak Prompts on Large Language Models](https://mlsanity.com/paper/2308.03825/) (2023)
- [Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks](https://mlsanity.com/paper/2404.02151/) (2024)
- [LMSYS-Chat-1M: A Large-Scale Real-World LLM Conversation Dataset](https://mlsanity.com/paper/2309.11998/) (2023)
- [Cognitive Behaviors that Enable Self-Improving Reasoners, or, Four Habits of Highly Effective STaRs](https://mlsanity.com/paper/2503.01307/) (2025)
- [CodeRL: Mastering Code Generation through Pretrained Models and Deep Reinforcement Learning](https://mlsanity.com/paper/2207.01780/) (2022)
- [Deliberative Alignment: Reasoning Enables Safer Language Models](https://mlsanity.com/paper/2412.16339/) (2024)
- [Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks](https://mlsanity.com/paper/1909.03496/) (2019)

## Built on by (most cited first, 18 of 18)

- [A Comprehensive Survey on Benchmarks and Solutions in Software Engineering of LLM-Empowered Agentic System](https://mlsanity.com/paper/2510.09721/) (2025)
- [SecureVibeBench: Benchmarking Secure Vibe Coding of AI Agents via Reconstructing Vulnerability-Introducing Scenarios](https://mlsanity.com/paper/2509.22097/) (2025)
- [Controlled Self-Evolution for Algorithmic Code Optimization](https://mlsanity.com/paper/2601.07348/) (2026)
- [CoRe3D: Collaborative Reasoning as a Foundation for 3D Intelligence](https://mlsanity.com/paper/2512.12768/) (2025)
- [Secure Code Generation via Online Reinforcement Learning with Vulnerability Reward Model](https://mlsanity.com/paper/2602.07422/) (2026)
- [CTIConnect: A Benchmark for Retrieval-Augmented LLMs over Heterogeneous Cyber Threat Intelligence](https://mlsanity.com/paper/2510.11974/) (2025)
- [From SFT to RL: Demystifying the Post-Training Pipeline for LLM-based Vulnerability Detection](https://mlsanity.com/paper/2602.14012/) (2026)
- [Learning to Generate Secure Code via Token-Level Rewards](https://mlsanity.com/paper/2602.23407/) (2026)
- [DualGauge: Automated Joint Security-Functionality Benchmarking of Specification-Only Code Generation by LLMs and Coding Agents](https://mlsanity.com/paper/2511.20709/) (2025)
- [SecPI: Secure Code Generation with Reasoning Models via Security Reasoning Internalization](https://mlsanity.com/paper/2604.03587/) (2026)
- [Adversarial Arena: Crowdsourcing Data Generation through Interactive Competition](https://mlsanity.com/paper/2604.17803/) (2026)
- [Themis: Training Robust Multilingual Code Reward Models for Flexible Multi-Criteria Scoring](https://mlsanity.com/paper/2605.00754/) (2026)
- [Amazon Nova AI Challenge -- Trusted AI: Advancing secure, AI-assisted software development](https://mlsanity.com/paper/2508.10108/) (2025)
- [ARuleCon: Agentic Security Rule Conversion](https://mlsanity.com/paper/2604.06762/) (2026)
- [SoK: AI Secure Code Generation: Progress, Pitfalls, and Paths Forward](https://mlsanity.com/paper/2606.25195/) (2026)
- [MiST: Mid-Training LLMs for Cybersecurity](https://mlsanity.com/paper/2609.18496/) (2026)
- [SecureVibe: Making Vibe Coding More Secure](https://mlsanity.com/paper/2609.38606/) (2026)
- [Last Secure Code Benchmark: Is the Code You Vibe Secure?](https://mlsanity.com/paper/2610.14766/) (2026)

## Abstract (arXiv)

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.