DERED Enhances Third-Party Library Detection
DERED mitigates deceptive reuse for improved TPL detection in binaries.
Stories per day · 30 days
1 RSS
Relatedcwe2assembly1deceptive-reuse1function-retrieval1indirect-calls1interprocedural1llm-benchmark1neural-decompilation1numanchor1primevul1reverse-engineering1third-party-libraries1vuln-detection1vulnerability-detection1
10 stories, newest first · one card per story
Stories tagged #binary-analysis
ICALL-BENCH tests LLM interprocedural reasoning on real binaries
On 50 C/C++ programs, five models’ verifiable success ranged from 23% to 81%, and more information access did not reliably help the hardest cases.
VulnLens uses structured explanations for binary vuln detection
On a 526-function PrimeVul test set it reported 75.5% accuracy and 77.2% F1, ahead of DeepSeek V4 Pro and Gemini 2.5 Pro on those metrics.
VulDec pairs neural decompilation with binary vulnerability detection
Tuned LLM detectors reach macro F1 0.61 on a realistic test set with a 3% vulnerability rate.
NumAnchor grounds numeric operands for assembly language models
Numeric-like operands were 40.9% to 45.7% of occurrences; retrieval Recall@1 rose by up to 35.0% without changing the host model.
Reverse engineer anything with agents, from app behavior down to native binaries
GitHub project rea uses AI agents to reverse engineer apps from behavior down to native binaries.
SoK: Are LLMs Reliable at Source Code Recovery? A Taxonomy and Empirical Evaluation
An SoK evaluates LLM binary-to-source recovery on 45,000 samples across architectures and languages.
DARPA Selects Xint to Use AI in Securing Military Messaging Apps
DARPA selected AIxCC winner Xint to autonomously analyze and patch vulnerabilities in Department of War messaging apps like Signal.
SEMA-GUARD: Semantic and Graph-Based Vulnerability Detection in Assembly Code
SEMA-GUARD uses semantic analysis and graph neural networks to detect vulnerabilities in assembly code, achieving 85.1% accuracy on a Juliet-derived benchmark.
When LLM Decompilers Recompile More and Preserve Less
Researchers show LLM decompiler outputs can recompile yet diverge behaviorally, proposing the Decompile-Diverge fuzzing oracle to catch hidden changes.