arXiv subjects

Sima Arasteh

Publications and source records attributed to Sima Arasteh.

6 recordsLinked to original sources

arXiv2025

When Machine Learning Meets Vulnerability Discovery: Challenges and Lessons Learned

In recent years, machine learning has demonstrated impressive results in various fields, including software vulnerability detection. Nonetheless, using machine learning to identify software vulnerabilities presents new challenges, especially regarding the scale of data involved, which was not a factor in traditional methods. Consequently, in spite of the rise of new machine-learning-based approaches in that space, important shortcomings persist regarding their evaluation. First, researchers often fail to provide concrete statistics about their training datasets, such as the number of samples for each type of vulnerability. Moreover, many methods rely on training with semantically similar functions rather than directly on vulnerable programs. This leads to uncertainty about the suitability of the datasets currently used for training. Secondly, the choice of a model and the level of granularity at which models are trained also affect the effectiveness of such vulnerability discovery approaches. In this paper, we explore the challenges of applying machine learning to vulnerability discovery. We also share insights from our two previous research papers, Bin2vec and BinHunter, which could enhance future research in this field.

cs.CR↗

arXiv2025

Data Flows in You: Benchmarking and Improving Static Data-flow Analysis on Binary Executables

Data-flow analysis is a critical component of security research. Theoretically, accurate data-flow analysis in binary executables is an undecidable problem, due to complexities of binary code. Practically, many binary analysis engines offer some data-flow analysis capability, but we lack understanding of the accuracy of these analyses, and their limitations. We address this problem by introducing a labeled benchmark data set, including 215,072 microbenchmark test cases, mapping to 277,072 binary executables, created specifically to evaluate data-flow analysis implementations. Additionally, we augment our benchmark set with dynamically-discovered data flows from 6 real-world executables. Using our benchmark data set, we evaluate three state of the art data-flow analysis implementations, in angr, Ghidra and Miasm and discuss their very low accuracy and reasons behind it. We further propose three model extensions to static data-flow analysis that significantly improve accuracy, achieving almost perfect recall (0.99) and increasing precision from 0.13 to 0.32. Finally, we show that leveraging these model extensions in a vulnerability-discovery context leads to a tangible improvement in vulnerable instruction identification.

cs.CR↗

arXiv2025

BinPool: A Dataset of Vulnerabilities for Binary Security Analysis

The development of machine learning techniques for discovering software vulnerabilities relies fundamentally on the availability of appropriate datasets. The ideal dataset consists of a large and diverse collection of real-world vulnerabilities, paired so as to contain both vulnerable and patched versions of each program. Naturally, collecting such datasets is a laborious and time-consuming task. Within the specific domain of vulnerability discovery in binary code, previous datasets are either publicly unavailable, lack semantic diversity, involve artificially introduced vulnerabilities, or were collected using static analyzers, thereby themselves containing incorrectly labeled example programs. In this paper, we describe a new publicly available dataset which we dubbed Binpool, containing numerous samples of vulnerable versions of Debian packages across the years. The dataset was automatically curated, and contains both vulnerable and patched versions of each program, compiled at four different optimization levels. Overall, the dataset covers 603 distinct CVEs across 89 CWE classes, 162 Debian packages, and contains 6144 binaries. We argue that this dataset is suitable for evaluating a range of security analysis tools, including for vulnerability discovery, binary function similarity, and plagiarism detection.

cs.CR↗

arXiv2025

Trim My View: An LLM-Based Code Query System for Module Retrieval in Robotic Firmware

The software compilation process has a tendency to obscure the original design of the system and makes it difficult both to identify individual components and discern their purpose simply by examining the resulting binary code. Although decompilation techniques attempt to recover higher-level source code from the machine code in question, they are not fully able to restore the semantics of the original functions. Furthermore, binaries are often stripped of metadata, and this makes it challenging to reverse engineer complex binary software. In this paper we show how a combination of binary decomposition techniques, decompilation passes, and LLM-powered function summarization can be used to build an economical engine to identify modules in stripped binaries and associate them with high-level natural language descriptions. We instantiated this technique with three underlying open-source LLMs -- CodeQwen, DeepSeek-Coder and CodeStral -- and measured its effectiveness in identifying modules in robotics firmware. This experimental evaluation involved 467 modules from four devices from the ArduPilot software suite, and showed that CodeStral, the best-performing backend LLM, achieves an average F1-score of 0.68 with an online running time of just a handful of seconds.

cs.CR↗

arXiv2024

The Reversing Machine: Reconstructing Memory Assumptions

Existing anti-malware software and reverse engineering toolkits struggle with stealthy sub-OS rootkits due to limitations of run-time kernel-level monitoring. A malicious kernel-level driver can bypass OS-level anti-virus mechanisms easily. Although static analysis of such malware is possible, obfuscation and packing techniques complicate offline analysis. Moreover, current dynamic analyzers suffer from virtualization performance overhead and create detectable traces that allow modern malware to evade them. To address these issues, we present \textit{The Reversing Machine} (TRM), a new hypervisor-based memory introspection design for reverse engineering, reconstructing memory offsets, and fingerprinting evasive and obfuscated user-level and kernel-level malware. TRM proposes two novel techniques that enable efficient and transparent analysis of evasive malware: hooking a binary using suspended process creation for hypervisor-based memory introspection, and leveraging Mode-Based Execution Control (MBEC) to detect user/kernel mode transitions and memory access patterns. Unlike existing malware detection environments, TRM can extract full memory traces in user and kernel spaces and hook the entire target memory map to reconstruct arrays, structures within the operating system, and possible rootkits. We perform TRM-assisted reverse engineering of kernel-level structures and show that it can speed up manual reverse engineering by 75\% on average. We obfuscate known malware with the latest packing tools and successfully perform similarity detection. Furthermore, we demonstrate a real-world attack by deploying a modified rootkit onto a driver that bypasses state-of-the-art security auditing tools. We show that TRM can detect each threat and that, out of 24 state-of-the-art AV solutions, only TRM can detect the most advanced threats.

cs.CR↗

arXiv2021

Bin2vec: Learning Representations of Binary Executable Programs for Security Tasks

Tackling binary program analysis problems has traditionally implied manually defining rules and heuristics, a tedious and time-consuming task for human analysts. In order to improve automation and scalability, we propose an alternative direction based on distributed representations of binary programs with applicability to a number of downstream tasks. We introduce Bin2vec, a new approach leveraging Graph Convolutional Networks (GCN) along with computational program graphs in order to learn a high dimensional representation of binary executable programs. We demonstrate the versatility of this approach by using our representations to solve two semantically different binary analysis tasks - functional algorithm classification and vulnerability discovery. We compare the proposed approach to our own strong baseline as well as published results and demonstrate improvement over state-of-the-art methods for both tasks. We evaluated Bin2vec on 49191 binaries for the functional algorithm classification task, and on 30 different CWE-IDs including at least 100 CVE entries each for the vulnerability discovery task. We set a new state-of-the-art result by reducing the classification error by 40% compared to the source-code-based inst2vec approach, while working on binary code. For almost every vulnerability class in our dataset, our prediction accuracy is over 80% (and over 90% in multiple classes).

cs.CR↗