6月第三周
往期回顾:
2021年8月-10月
- 为什么机器学习解决网络安全问题总是失败:脆弱的系统工程https://toooold.com/2021/10/11/why_ml_fails_security_frag_cn.html
- 为什么机器学习解决网络安全问题总是失败:谈谈特征空间 https://toooold.com/2021/09/27/why_ml_fails_solving_sec_algo.html
- 【论文】 The mathematics of adversarial attacks in AI — Why deep learning is unstable despite the existence of stable neural networks https://arxiv.org/abs/2109.06098
- OpenAttack更新了 https://mp.weixin.qq.com/s/gAmoczeK3mUjZhL7D1EhdA
- 【论文】 APICRAFT: Fuzz Driver Generation for Closed-source SDK Libraries https://paper.seebug.org/1706/
- 【论文】An Empirical Cybersecurity Evaluation of GitHub Copilot’s Code Contributions https://arxiv.org/pdf/2108.09293.pdf
- 【论文】Identifying Non-Control Security-Critical Data in Program Binaries with a Deep Neural Model https://arxiv.org/pdf/2108.12071.pdf
- 在深度神经网络里隐藏任意恶意软件 https://toooold.com/2021/09/05/hide_malware_ann.html
- 基于编码注入的对抗性NLP攻击 https://arxiv.org/abs/2106.09898
- CTF中一些AI安全相关的题目分析 https://www.anquanke.com/post/id/251125
- GitHub Copilot的安全分析 https://www.fast.ai/2021/07/19/copilot/
- 用于评估各种“深度造假”检测模型的模块化框架FakeFinder https://github.com/IQTLabs/FakeFinder