Unmasking the Shadows: Pinpoint the Implementations of Anti-Dynamic Analysis Techniques in Malware Using LLM

TL;DR

This paper introduces an LLM-based workflow to identify anti-dynamic analysis techniques in malware code, aiding reverse engineers in debugging and improving malware detection robustness.

Contribution

It presents a novel application of large language models to locate anti-dynamic analysis code in malware, achieving high accuracy on known and real-world samples.

Findings

Identified 87.80% of known TADA implementations in public repositories.

Successfully pinpointed TADA locations in 4 real malware samples.

Demonstrated effectiveness in aiding reverse engineering processes.

Abstract

Sandboxes and other dynamic analysis processes are prevalent in malware detection systems nowadays to enhance the capability of detecting 0-day malware. Therefore, techniques of anti-dynamic analysis (TADA) are prevalent in modern malware samples, and sandboxes can suffer from false negatives and analysis failures when analyzing the samples with TADAs. In such cases, human reverse engineers will get involved in conducting dynamic analysis manually (i.e., debugging, patching), which in turn also gets obstructed by TADAs. In this work, we propose a Large Language Model (LLM) based workflow that can pinpoint the location of the TADA implementation in the code, to help reverse engineers place breakpoints used in debugging. Our evaluation shows that we successfully identified the locations of 87.80% known TADA implementations adopted from public repositories. In addition, we successfully pinpoint the locations of TADAs in 4 well-known malware samples that are documented in online malware analysis blogs.