Crowdstrike, Inc.

Alexandru Dinu

Mihaela-Petruta Gaman

Cristian Viorel Popa

Stefan-Bogdan Cocea

Paul Sumedrea

Damian Monea

Methods and systems for applying a diffusion model to adversarial purification and generating adversarial samples in malware detection are disclosed. According to an example, a malware file is inputted to a diffusion model to obtain an adversarial sample by altering content of the malware file. The adversarial sample is further tested by a malware detector. In some examples, the content of an input file may be encoded prior to be processed by the diffusion model. If the malware detector can identify the adversarial sample as a malware file, the diffusion model is updated to further alter the content until the adversarial sample successfully deceives the malware detector. According to another example, an executable file is purified using a diffusion model prior to be inputted to a malware detector. The diffusion model may remove potential malware content from the executable file, thus improving the performance of the malware detector.

Applying diffusion models in adversarial purification and adversarial sample generation in cybersecurity

An approach is provided that provides a plurality of source code samples to an artificial intelligence model (AIM) trained to describe source code based on performing semantic analysis on the source code. The approach produces, using the AIM, a plurality of semantic descriptions that describe the plurality of source code samples. Then, the approach converts the plurality of semantic descriptions into a plurality of semantic embeddings. In turn, the approach creates a plurality of clusters from the plurality of semantic embeddings, wherein each one of the plurality of clusters corresponds to two or more of the plurality of source code samples.

Secure code clustering through LLM-based semantic analysis

The present disclosure provides an approach of providing, to an artificial intelligence (AI) model, a malicious script that includes a malicious behavior. The AI model is configured to modify software code of the malicious script to produce modified software code that obfuscates the malicious behavior. The approach produces, by a processing device using the AI model, an adversarial script that includes the modified software code that obfuscates the malicious behavior. In turn, the approach initiates a malware detector to test the adversarial script.

Growing community of inventors

Bucharest, Romania

Alexandru Dinu