Hodaya Binyamini

Accenture Labs
Automatic Attack Flow Extraction from Attack Textual Report
Hodaya Binyamin

Bio

Hodaya is an applied researcher at Accenture Labs, where she leverages machine learning (ML) and deep learning (DL) to develop innovative solutions for real-world challenges. Her current research focuses on using NLP and knowledge graphs to enhance cybersecurity threat detection and response. Hodaya holds an MSc degree in Information Systems Engineering and a BSc degree in Communication Systems Engineering, both from Ben-Gurion University. She is passionate about using her expertise to make a positive impact on the world.

Abstract

Cybersecurity researchers spend countless hours manually documenting complex, multi-step attacks in textual reports. These reports are crucial for knowledge sharing and analysis. Usually, the attacks are complex and contain more than one step, as well as conditions between those steps. To facilitate this communication, MITRE came up with “Attack Flow”, a schema that describes all the necessary attack steps, conditions, and outcomes. Representing attacks as graphs brings two key benefits: enhanced visualization for attack analysts and streamlined comparison for validating environment safety.

This talk presents a novel approach using Large Language Models (LLMs) to automate the extraction and construction of Attack Flow from textual attack reports. Our method leverages the power of LLMs to accurately identify attack steps, conditions, and their relationships, transforming unstructured reports into structured attack flow representations. This automation can significantly improve report clarity, expedite knowledge sharing, and facilitate deeper analysis of attack patterns.

Agenda

8:45 Reception
9:30 Opening remarks by WiDS TLV ambassadors Noah Eyal Altman, Or Basson, and Nitzan Gado
9:45 Dr. Aya Soffer, IBM: "Putting Generative AI to Work: What Have We Learned So Far?"
10:15 Prof. Reut Tsarfaty, Bar-llan University: "Will Hebrew Speakers Be Able to Use Generative AI in Their Native Tongue?"
10:45 Poster Pitches
10:55 Break
11:10 Lightning talks
12:30 Lunch & poster session
13:30 Roundtable session & poster session
14:15 Roundtable closing
14:30 Break
14:40 Naomi Ken Korem, Lightricks: "Mastering the Art of Generative Models: Training and Controlling Text-to-Video Models"
15:00 Dr. Yael Mathov, Intuit: "Surviving the AI-pocalypse: Your Guide to LLM Security"
15:20 Closing remarks
15:30 The end