MAGMA combines RAG with a stochastic consistency ensemble over dual code embeddings to derive Function Evidence Strength and Evidence Conflict Score metrics, enabling reject-option decisions and achieving 98.4% malware detection.
AsmRAG: LLM-Driven Malware Detection by Retrieving Functionally Similar Assembly Code
1 Pith paper cite this work. Polarity classification is still indexing.
abstract
Deep learning malware detectors achieve high classification accuracy but suffer from severe interpretability limitations, typically returning probabilistic verdicts that lack forensic context. We introduce AsmRAG, a framework performing malware analysis through Assembly-Level Retrieval-Augmented Generation. Unlike classifiers built on global statistical features, AsmRAG reformulates detection as an evidence-based retrieval task. The system uses a code-specialized Large Language Model (LLM) to analyze assembly functions and convert them into semantic embeddings. This process constructs a searchable knowledge base resilient to syntactic obfuscation. For inference, we propose a Density-Weighted Anchor Selection mechanism that isolates the primary unit of malicious logic within a binary to extract verifiable forensic evidence and resist evasion attempts. Testing on a curated dataset of 40k binaries shows AsmRAG reaching a detection F1-score of 96% alongside a family attribution F1-score of 95%. Comparisons confirm this semantic retrieval approach remains robust against metamorphic obfuscation. When holistic baselines (EMBER and ResNeXt) degrade, our methodology gives Security Operations Centers a transparent and reliable alternative.
citation-role summary
citation-polarity summary
fields
cs.CR 1years
2026 1verdicts
UNVERDICTED 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Quantifiable Uncertainty: A Stochastic Consensus Multi-Agent RAG Framework for Robust Malware Detection
MAGMA combines RAG with a stochastic consistency ensemble over dual code embeddings to derive Function Evidence Strength and Evidence Conflict Score metrics, enabling reject-option decisions and achieving 98.4% malware detection.