惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

美团技术团队
Cyber Security Advisories - MS-ISAC
Cyber Security Advisories - MS-ISAC
Martin Fowler
Martin Fowler
雷峰网
雷峰网
IT之家
IT之家
小众软件
小众软件
M
MIT News - Artificial intelligence
博客园 - 聂微东
J
Java Code Geeks
Blog — PlanetScale
Blog — PlanetScale
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
A
About on SuperTechFans
G
Google Developers Blog
Engineering at Meta
Engineering at Meta
Recent Announcements
Recent Announcements
freeCodeCamp Programming Tutorials: Python, JavaScript, Git & More
The GitHub Blog
The GitHub Blog
F
Fortinet All Blogs
C
Check Point Blog
云风的 BLOG
云风的 BLOG
腾讯CDC
H
Help Net Security
Y
Y Combinator Blog
I
InfoQ

cs.CR updates on arXiv.org

On the Security of Research Artifacts SafeHarbor: Hierarchical Memory-Augmented Guardrail for LLM Agent Safety Agentic Vulnerability Reasoning on Windows COM Binaries From Beats to Breaches:How Offensive AI Infers Sensitive User Information from Playlists Undetectable Backdoors in Model Parameters: Hiding Sparse Secrets in High Dimensions When Embedding-Based Defenses Fail: Rethinking Safety in LLM-Based Multi-Agent Systems Token-Efficient Change Detection in LLM APIs Selfie-Capture Dynamics as an Auxiliary Signal Against Deepfakes and Injection Attacks for Mobile Identity Verification Trident: Improving Malware Detection with LLMs and Behavioral Features When Alignment Isn't Enough: Response-Path Attacks on LLM Agents RefusalGuard: Geometry-Preserving Fine-Tuning for Safety in LLMs Checkerboard: A Simple, Effective, Efficient and Learning-free Clean Label Backdoor Attack with Low Poisoning Budget Block-wise Codeword Embedding for Reliable Multi-bit Text Watermarking Secret Stealing Attacks on Local LLM Fine-Tuning through Supply-Chain Model Code Backdoors Enhancing Linux Privilege Escalation Attack Capabilities of Local LLM Agents Defusing the Trigger: Plug-and-Play Defense for Backdoored LLMs via Tail-Risk Intrinsic Geometric Smoothing Evaluating Jailbreaking Vulnerabilities in LLMs Deployed as Assistants for Smart Grid Operations: A Benchmark Against NERC Standards Behavioral Canaries: Auditing Private Retrieved Context Usage in RL Fine-Tuning FlexServe: A Fast and Secure LLM Serving System for Mobile Devices with Flexible Resource Isolation Breaking MCP with Function Hijacking Attacks: Novel Threats for Function Calling and Agentic Models Text Steganography with Dynamic Codebook and Multimodal Large Language Model TwoHamsters: Benchmarking Multi-Concept Compositional Unsafety in Text-to-Image Models Fundamental Limitations of Favorable Privacy-Utility Guarantees for DP-SGD Symbolic Guardrails for Domain-Specific Agents: Stronger Safety and Security Guarantees Without Sacrificing Utility Hardening x402: PII-Safe Agentic Payments via Pre-Execution Metadata Filtering QShield: Securing Neural Networks Against Adversarial Attacks using Quantum Circuits Hijacking Text Heritage: Hiding the Human Signature through Homoglyphic Substitution Like a Hammer, It Can Build, It Can Break: Large Language Model Uses, Perceptions, and Adoption in Cybersecurity Operations on Reddit Private Seeds, Public LLMs: Realistic and Privacy-Preserving Synthetic Data Generation One Word at a Time: Incremental Completion Decomposition Breaks LLM Safety
Learn from Your Mistakes: Tree-like Self-Play for Secure ...
Wenqi Chen, Ziyan Zhang, Bing Wang, Lin Liu, Hengheng Zhang, Zhe · 2026-06-02 · via cs.CR updates on arXiv.org

While Large Language Models (LLMs) excel in code generation, they remain prone to replicating subtle yet critical vulnerabilities endemic to their training data. Current alignment techniques, such as Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL), typically apply coarse-grained optimization at the sequence level. This approach often fails to address the localized nature of security flaws, where a single incorrect token choice can compromise an entire program. To bridge this gap, we introduce Tree-like Self-Play (TSP), a framework that reframes secure code generation as a fine-grained sequential decision process. Unlike standard methods that blindly maximize likelihood, TSP constructs a decision tree where the model explores branching trajectories--generating both secure "golden paths" and vulnerable variants. By treating code generation as a self-play game, the model learns to strictly discriminate against its own localized errors. This provides a dense, on-policy learning signal that forces self-correction precisely at the critical decision nodes where vulnerabilities typically emerge. Our experiments demonstrate that TSP fundamentally enhances model reliability. In Python security benchmarks, TSP boosts CodeLlama-7B's pass rate (SPR@1) to 75.8%, significantly outperforming SFT (57.0%) and unstructured self-play baselines. Crucially, TSP induces robust out-of-distribution generalization: the model not only reduces vulnerabilities in unseen categories (CWEs) by 24.5% but also successfully transfers security principles learned from C/C++ to diverse languages, including Python, Go, and JavaScript. This suggests that TSP does not merely memorize patches, but internalizes abstract, language-agnostic security logic.