LLM Safety States: Aligned vs. Abliterated in Software Security
Examining how aligned and abliterated large language models fare in software security tasks like vulnerability detection and patch validation. key insights into model safety states and their…
Read the full articleYou might also wanna read
Security Risks of Malicious Backdoors in Large Language Models
LLM security is a critical risk for open-weight models. Learn how malicious backdoors are easily fine-tuned into AI agents to execute harmfu
pub.aimind.so·11mo ago
Open-Source LLM Safety Vulnerabilities: How Chat Template Formatting Gates Alignment in Models Like Gemma and Qwen
How a Single Function Call Gates Safety Alignment in Gemma, Qwen, and Other Open-Source LLMs
teendifferent.substack.com·5mo ago
BPE Tokenization Creates Exploitable Safety Gaps in LLM Alignment, Study Finds
Character-level perturbations bypass safety alignment in modern LLMs despite leaving prompts human-readable. We identify and test a central
arxiv.org·7d ago
Reflections on LLMs and Their Impact on Software Development Practices
a short post
LLMs Detect Vulnerabilities by Recognizing Safe Code Patterns, Not Vulnerable Ones, Study Finds
Large language models (LLMs) can detect software vulnerabilities, but how do they actually identify vulnerable code? We address this questio
arxiv.org·15d ago
Benchmarking Frontier LLMs on Real-World CVE Patching: Mixed Results and Methodological Challenges
Benchmarking LLMs on real-world CVE patching
Comments
Sign in to join the conversation.
No comments yet. Be the first.