PPhying中文
Curated security research, vulnerabilities, advisories and tools for practitioners.

Vulnerability Discovery with LLM-Powered Patch Diffing

bishopfox.com | research | #ai-security | #vulnerability-research | #automation | #binary-analysis | #llm | #patch-diffing
📄 Raw .md⬇️ Download .md🔗 Original ↗

Summary

Bishop Fox tested LLM-assisted patch diffing on four CVSS 9.4+ vulnerabilities: decompiled diffs fed to Claude Sonnet 3.7, which ranked changed functions by advisory relevance, cutting analysis time.
Published
Collected

Related coverage

research · vercel.com
DeepsecBench: evaluating model performance in finding cybersecurity vulnerabilities
Today we're releasing DeepsecBench, a benchmark that evaluates how well different models find security vulnerabilities in application code.
blog · hackenproof.com
How We Built an Autonomous AI Security Triager: From LLM to 20,000+ Real Reports
AI made bug hunters sharper, and report volume doubled. How HackenProof built an autonomous AI triager that now handles 70% of submissions.
research · blog.quarkslab.com
State of Post-Quantum Cryptography implementations
In this blogpost, we first review the landscape of Post-Quantum Cryptography implementations across some of the most popular cryptographic libraries. Following that, we present the results of a differential fuzzing campaign on few of the presented implementations.
research · blog.quarkslab.com
Extending LLVM's BOLT-based Binary Analyser to Validate Stack Variable Initialisation
Quarkslab extended LLVM's BOLT-based binary analyser for OSTIF with a scanner that validates -ftrivial-auto-var-init at the binary level, verifying stack-variable initialisation hardening.
research · blog.quarkslab.com
QBDI vs TritonDSE against a VM: who will be the fastest?
Quarkslab solves a custom-VM CTF challenge twice — with symbolic execution engine TritonDSE and instrumentation framework QBDI — and compares which approach gets to the solution fastest.
research · blog.quarkslab.com
Introducing TritonDSE: A framework for dynamic symbolic execution in Python
Introducing TritonDSE, a Python library atop Triton for customizable dynamic symbolic execution: high-level program exploration primitives, hooks, and the base of PASTIS's white-box fuzzer.
Back