arXiv subjects

Preston K. Robinette

Publications and source records attributed to Preston K. Robinette.

7 recordsLinked to original sources

arXiv2026

Towards Verified and Targeted Explanations through Formal Methods

As deep neural networks are deployed in safety-critical domains such as autonomous driving and medical diagnosis, stakeholders need explanations that are interpretable but also trustworthy with formal guarantees. Existing XAI methods fall short: heuristic attribution techniques (e.g., LIME, Integrated Gradients) highlight influential features but offer no mathematical guarantees about decision boundaries, while formal methods verify robustness yet remain untargeted, analyzing the nearest boundary regardless of whether it represents a critical risk. In safety-critical systems, not all misclassifications carry equal consequences; confusing a "Stop" sign for a "60 kph" sign is far more dangerous than confusing it with a "No Passing" sign. We introduce ViTaX (Verified and Targeted Explanations), a formal XAI framework that generates targeted semifactual explanations with mathematical guarantees. For a given input (class y) and a user-specified critical alternative (class t), ViTaX: (1) identifies the minimal feature subset most sensitive to the y->t transition, and (2) applies formal reachability analysis to guarantee that perturbing these features by epsilon cannot flip the classification to t. We formalize this through Targeted epsilon-Robustness, certifying whether a feature subset remains robust under perturbation toward a specific target class. ViTaX is the first method to provide formally guaranteed explanations of a model's resilience against user-identified alternatives. Evaluations on MNIST, GTSRB, EMNIST, and TaxiNet demonstrate over 30% fidelity improvement with minimal explanation cardinality.

cs.LG↗

arXiv2026

SUGAR: A Sweeter Spot for Generative Unlearning of Many Identities

Recent advances in 3D-aware generative models have enabled high-fidelity image synthesis of human identities. However, this progress raises urgent questions around user consent and the ability to remove specific individuals from a model's output space. We address this by introducing SUGAR, a framework for scalable generative unlearning that enables the removal of many identities (simultaneously or sequentially) without retraining the entire model. Rather than projecting unwanted identities to unrealistic outputs or relying on static template faces, SUGAR learns a personalized surrogate latent for each identity, diverting reconstructions to visually coherent alternatives while preserving the model's quality and diversity. We further introduce a continual utility preservation objective that guards against degradation as more identities are forgotten. SUGAR achieves state-of-the-art performance in removing up to 200 identities, while delivering up to a 700% improvement in retention utility compared to existing baselines. Our code is publicly available at https://github.com/judydnguyen/SUGAR-Generative-Unlearn.

cs.CV↗

arXiv2025

Sanitizing Hidden Information with Diffusion Models

Information hiding is the process of embedding data within another form of data, often to conceal its existence or prevent unauthorized access. This process is commonly used in various forms of secure communications (steganography) that can be used by bad actors to propagate malware, exfiltrate victim data, and discreetly communicate. Recent work has utilized deep neural networks to remove this hidden information in a defense mechanism known as sanitization. Previous deep learning works, however, are unable to scale efficiently beyond the MNIST dataset. In this work, we present a novel sanitization method called DM-SUDS that utilizes a diffusion model framework to sanitize/remove hidden information from image-into-image universal and dependent steganography from CIFAR-10 and ImageNet datasets. We evaluate DM-SUDS against three different baselines using MSE, PSNR, SSIM, and NCC metrics and provide further detailed analysis through an ablation study. DM-SUDS outperforms all three baselines and significantly improves image preservation MSE by 50.44%, PSNR by 12.69%, SSIM by 11.49%, and NCC by 3.26% compared to previous deep learning approaches. Additionally, we introduce a novel evaluation specification that considers the successful removal of hidden information (safety) as well as the resulting quality of the sanitized image (utility). We further demonstrate the versatility of this method with an application in an audio case study, demonstrating its broad applicability to additional domains.

cs.CR↗

arXiv2025

Blind Visible Watermark Removal with Morphological Dilation

Visible watermarks pose significant challenges for image restoration techniques, especially when the target background is unknown. Toward this end, we present MorphoMod, a novel method for automated visible watermark removal that operates in a blind setting -- without requiring target images. Unlike existing methods, MorphoMod effectively removes opaque and transparent watermarks while preserving semantic content, making it well-suited for real-world applications. Evaluations on benchmark datasets, including the Colored Large-scale Watermark Dataset (CLWD), LOGO-series, and the newly introduced Alpha1 datasets, demonstrate that MorphoMod achieves up to a 50.8% improvement in watermark removal effectiveness compared to state-of-the-art methods. Ablation studies highlight the impact of prompts used for inpainting, pre-removal filling strategies, and inpainting model performance on watermark removal. Additionally, a case study on steganographic disorientation reveals broader applications for watermark removal in disrupting high-level hidden messages. MorphoMod offers a robust, adaptable solution for watermark removal and opens avenues for further advancements in image restoration and adversarial manipulation.

cs.CV↗

arXiv2024

Trigger-Based Fragile Model Watermarking for Image Transformation Networks

In fragile watermarking, a sensitive watermark is embedded in an object in a manner such that the watermark breaks upon tampering. This fragile process can be used to ensure the integrity and source of watermarked objects. While fragile watermarking for model integrity has been studied in classification models, image transformation/generation models have yet to be explored. We introduce a novel, trigger-based fragile model watermarking system for image transformation/generation networks that takes advantage of properties inherent to image outputs. For example, manifesting watermarks as specific visual patterns, styles, or anomalies in the generated content when particular trigger inputs are used. Our approach, distinct from robust watermarking, effectively verifies the model's source and integrity across various datasets and attacks, outperforming baselines by 94%. We conduct additional experiments to analyze the security of this approach, the flexibility of the trigger and resulting watermark, and the sensitivity of the watermarking loss on performance. We also demonstrate the applicability of this approach on two different tasks (1 immediate task and 1 downstream task). This is the first work to consider fragile model watermarking for image transformation/generation networks.

cs.CR↗

arXiv2024

Case Study: Neural Network Malware Detection Verification for Feature and Image Datasets

Malware, or software designed with harmful intent, is an ever-evolving threat that can have drastic effects on both individuals and institutions. Neural network malware classification systems are key tools for combating these threats but are vulnerable to adversarial machine learning attacks. These attacks perturb input data to cause misclassification, bypassing protective systems. Existing defenses often rely on enhancing the training process, thereby increasing the model's robustness to these perturbations, which is quantified using verification. While training improvements are necessary, we propose focusing on the verification process used to evaluate improvements to training. As such, we present a case study that evaluates a novel verification domain that will help to ensure tangible safeguards against adversaries and provide a more reliable means of evaluating the robustness and effectiveness of anti-malware systems. To do so, we describe malware classification and two types of common malware datasets (feature and image datasets), demonstrate the certified robustness accuracy of malware classifiers using the Neural Network Verification (NNV) and Neural Network Enumeration (nnenum) tools, and outline the challenges and future considerations necessary for the improvement and refinement of the verification of malware classification. By evaluating this novel domain as a case study, we hope to increase its visibility, encourage further research and scrutiny, and ultimately enhance the resilience of digital systems against malicious attacks.

cs.CR↗

arXiv2023

SUDS: Sanitizing Universal and Dependent Steganography

Steganography, or hiding messages in plain sight, is a form of information hiding that is most commonly used for covert communication. As modern steganographic mediums include images, text, audio, and video, this communication method is being increasingly used by bad actors to propagate malware, exfiltrate data, and discreetly communicate. Current protection mechanisms rely upon steganalysis, or the detection of steganography, but these approaches are dependent upon prior knowledge, such as steganographic signatures from publicly available tools and statistical knowledge about known hiding methods. These dependencies render steganalysis useless against new or unique hiding methods, which are becoming increasingly common with the application of deep learning models. To mitigate the shortcomings of steganalysis, this work focuses on a deep learning sanitization technique called SUDS that is not reliant upon knowledge of steganographic hiding techniques and is able to sanitize universal and dependent steganography. SUDS is tested using least significant bit method (LSB), dependent deep hiding (DDH), and universal deep hiding (UDH). We demonstrate the capabilities and limitations of SUDS by answering five research questions, including baseline comparisons and an ablation study. Additionally, we apply SUDS to a real-world scenario, where it is able to increase the resistance of a poisoned classifier against attacks by 1375%.

cs.CR↗