AXRP Episode 38.8 - David Duvenaud on Sabotage Evaluations and the Post-AGI Future
"对抗性攻击" 相关文章
OpenAI新研究：o1增加推理时间就能防攻击，网友：DeepSeek也受益
Strengthening Security Throughout the ML/AI Lifecycle
用“自动化红队测试”解决AI越狱问题，Haize Labs创业7个月估值一亿美元
Imposter.AI: Unveiling Adversarial Attack Strategies to Expose Vulnerabilities in Advanced Large Language Models
击败人类又怎样？“超人”AI简直不堪一击？研究发现：ChatGPT等大模型也不行
Advancing Robustness in Neural Information Retrieval: A Comprehensive Survey and Benchmarking Framework
击败人类又怎样？“超人”AI简直不堪一击？研究发现：ChatGPT等大模型也不行
This AI Paper from the National University of Singapore Introduces a Defense Against Adversarial Attacks on LLMs Utilizing Self-Evaluation
MALT (Mesoscopic Almost Linearity Targeting): A Novel Adversarial Targeting Method based on Medium-Scale Almost Linearity Assumptions
Safeguarding Healthcare AI: Exposing and Addressing LLM Manipulation Risks
WildTeaming: An Automatic Red-Team Framework to Compose Human-like Adversarial Attacks Using Diverse Jailbreak Tactics Devised by Creative and Self-Motivated Users in-the-Wild
多模态大语言模型的致命漏洞：语音攻击
Model Explainability Forum - #401
Attacking Malware with Adversarial Machine Learning, w/ Edward Raff - #529
