arXiv subjects

Oleg Y. Rogov

Publications and source records attributed to Oleg Y. Rogov.

At least 19 recordsLinked to original sources

arXiv2026

Large Audio Language Models for Spoofing-Aware Speaker Verification

Recent advances in text-to-speech and voice cloning make high-quality spoofing inexpensive and scalable, threatening voice authentication systems, especially automatic speaker verification (ASV). Existing defenses mainly address this threat through binary countermeasures (CMs) for deepfake detection or spoofing-aware speaker verification (SASV), where current systems are dominated by modular ASV-CM fusion and cascaded pipelines. Although large audio language models (LALMs) have shown promise on related audio tasks, including CM and ASV, their use for SASV remains unexplored, despite their capacity to produce natural-language rationales for auditing and robustness beyond discriminative predictions. This work systematically evaluates LALMs for SASV against conventional pipelines under zero-shot prompting, supervised adaptation, reasoning-oriented training, and reinforcement-learning-based optimization. Our results show that pretrained LALMs are near chance in the zero-shot setting, confirming that they are not natively suited to SASV, but that task-specific adaptation closes this gap. We further find that competitive SASV performance can be achieved through several distinct routes. These findings position LALMs as a promising and auditable foundation for unified SASV, while clarifying where conventional cascade systems still lead.

cs.SD↗

arXiv2026

Towards Robust Speech Deepfake Detection via Human-Inspired Reasoning

The modern generative audio models can be used by an adversary in an unlawful manner, specifically, to impersonate other people to gain access to private information. To mitigate this issue, speech deepfake detection (SDD) methods started to evolve. Unfortunately, current SDD methods generally suffer from the lack of generalization to new audio domains and generators. More than that, they lack interpretability, especially human-like reasoning that would naturally explain the attribution of a given audio to the bona fide or spoof class and provide human-perceptible cues. In this paper, we propose HIR-SDD, a novel SDD framework that combines the strengths of Large Audio Language Models (LALMs) with the chain-of-thought reasoning derived from the novel proposed human-annotated dataset. Experimental evaluation demonstrates both the effectiveness of the proposed method and its ability to provide reasonable justifications for predictions.

cs.SD↗

arXiv2026

LLM-Guided Program Evolution for Targeted Black-Box Attacks on Perceptual Hash Algorithms

Perceptual hash algorithms (PHAs) are widely deployed to detect image forgery under benign transformations, yet their robustness against adversarially chosen perturbations remains poorly understood and rarely comes with provable guarantees. We propose a novel evolutionary framework based on GigaEvo and OpenEvolve for targeted second-image attacks on perceptual hash algorithms. We assess attack performance using a composite score that jointly accounts for the fraction of adversarial images whose normalized Hamming distance to the target hash falls below threshold p (Attack Success Rate), the number of queries issued to the hash function, and the L2 distortion relative to the original image. Experiments on four deployed PHAs (pHash, PDQ, PhotoDNA, NeuralHash) across 30 ImageNet image pairs demonstrate that our evolutionary approach achieves comparable or better ASR than existing black-box baselines using substantially fewer queries to the hash function, while simultaneously producing adversarial images with lower L2 distortion relative to the originals. The best evolved programs reduce the pre-defined composite attack score relative to the best optimized seed by 41.2% for NeuralHash, 38.3% for PDQ, 34.0% for pHash, and 8.1% for PhotoDNA. Unlike gradient-based methods, our framework requires no internal knowledge of PHA architectures and naturally handles the non-differentiable, discretized nature of hash outputs. These results reveal previously unreported vulnerabilities in widely deployed content-moderation pipelines and motivate the development of provably robust perceptual hashing 1schemes.

cs.CR↗

arXiv2026

AASIST3: KAN-Enhanced AASIST Speech Deepfake Detection using SSL Features and Additional Regularization for the ASVspoof 2024 Challenge

Automatic Speaker Verification (ASV) systems, which identify speakers based on their voice characteristics, have numerous applications, such as user authentication in financial transactions, exclusive access control in smart devices, and forensic fraud detection. However, the advancement of deep learning algorithms has enabled the generation of synthetic audio through Text-to-Speech (TTS) and Voice Conversion (VC) systems, exposing ASV systems to potential vulnerabilities. To counteract this, we propose a novel architecture named AASIST3. By enhancing the existing AASIST framework with Kolmogorov-Arnold networks, additional layers, encoders, and pre-emphasis techniques, AASIST3 achieves a more than twofold improvement in performance. It demonstrates minDCF results of 0.5357 in the closed condition and 0.1414 in the open condition, significantly enhancing the detection of synthetic voices and improving ASV security. \textbf{The new version of the model is publicly available at \href{https://huggingface.co/lab260/Spectra-AASIST3}{\underline{HuggingFace (2026)}}}

cs.SD↗

arXiv2026

DIPLI: Deep Image Prior Lucky Imaging for Blind Astronomical Image Restoration

Modern image restoration and super-resolution methods utilize deep learning due to its superior performance compared to traditional algorithms. However, deep learning typically requires large labeled training datasets, which are rarely available in astrophotography. Deep Image Prior (DIP) bypasses this constraint by performing unsupervised optimization on a single image without training data; however, DIP often suffers from overfitting, artifact generation, and instability. This work proposes DIPLI - a framework designed specifically for resolved, high-contrast astronomical targets that shifts from single-frame to multi-frame processing using the Back Projection technique, combined with dense optical flow estimation via the TVNet model, and replaces deterministic predictions with Monte Carlo estimation obtained through Stochastic Gradient Langevin Dynamics (SGLD). A comprehensive evaluation compares the method against the original DIP, the transformer-based model RVRT, and the diffusion-based model DiffIR2VR-Zero on synthetic data with ground truth, while comparing qualitatively against Lucky Imaging on real astronomical data. On synthetic datasets, DIPLI achieves the best perceptual fidelity scores (LPIPS in 12/12 and DISTS in 10/12 scenarios), while the diffusion-based DiffIR2VR-Zero achieves the best pixel-level distortion scores (PSNR in 9/12 and SSIM in 8/12 scenarios), consistent with the well-known perceptual-distortion trade-off in image restoration. Compared to classical Lucky Imaging, the model requires far fewer input frames (7-13 versus thousands) and avoids the need for early stopping that limits standard DIP. Qualitative evaluation on real-world data of resolved solar-system objects, where ground truth is unavailable and domain shifts typically hinder generalization, suggests that the method appears to preserve fine detail while suppressing noise and artifacts.

cs.CV↗

arXiv2026

LLM-Guided Prompt Evolution for Password Guessing

Passwords still remain a dominant authentication method, yet their security is routinely subverted by predictable user choices and large-scale credential leaks. Automated password guessing is a key tool for stress-testing password policies and modeling attacker behavior. This paper applies LLM-driven evolutionary computation to automatically optimize prompts for the LLM password guessing framework. Using OpenEvolve, an open-source system combining MAP-Elites quality-diversity search with an island population model we evolve prompts that maximize cracking rate on a RockYou-derived test set. We evaluate three configurations: a local setup with Qwen3 8B, a single compact cloud model Gemini-2.5 Flash, and a two-model ensemble of frontier LLMs. The approach raises the cracking rates from 2.02\% to 8.48\%. Character distribution analysis further confirms how evolved prompts produce statistically more realistic passwords. Automated prompt evolution is a low-barrier yet effective way to strengthen LLM-based password auditing and underlining how attack pipelines show tendency via automated improvements.

cs.CR↗

arXiv2026

Probabilistic Verification of Voice Anti-Spoofing Models

Recent advances in generative models have amplified the risk of malicious misuse of speech synthesis technologies, enabling adversaries to impersonate target speakers and access sensitive resources. Although speech deepfake detection has progressed rapidly, most existing countermeasures lack formal robustness guarantees or fail to generalize to unseen generation techniques. We propose PV-VASM, a probabilistic framework for verifying the robustness of voice anti-spoofing models (VASMs). PV-VASM estimates the probability of misclassification under text-to-speech (TTS), voice cloning (VC), and parametric signal transformations. The approach is model-agnostic and enables robustness verification against unseen speech synthesis techniques and input perturbations. We derive a theoretical upper bound on the error probability and validate the method across diverse experimental settings, demonstrating its effectiveness as a practical robustness verification tool.

cs.SD↗

arXiv2026

Speech-to-LaTeX: New Models and Datasets for Converting Spoken Equations and Sentences

Conversion of spoken mathematical expressions is a challenging task that involves transcribing speech into a strictly structured symbolic representation while addressing the ambiguity inherent in the pronunciation of equations. Although significant progress has been achieved in automatic speech recognition (ASR) and language models (LM), the problem of converting spoken mathematics into LaTeX remains underexplored. This task directly applies to educational and research domains, such as lecture transcription or note creation. Based on ASR post-correction, prior work requires 2 transcriptions, focuses only on isolated equations, has a limited test set, and provides neither training data nor multilingual coverage. To address these issues, we present the first fully open-source large-scale dataset, comprising over 66,000 human-annotated audio samples of mathematical equations and sentences in English and Russian, drawn from diverse scientific domains. In addition to the ASR post-correction models and few-shot prompting, we apply audio language models, demonstrating comparable character error rate (CER) results on the MathSpeech benchmark (28% vs. 30%) for the equations conversion. In contrast, on the proposed S2L-equations benchmark, our models outperform the MathSpeech model by a substantial margin of more than 36 percentage points, even after accounting for LaTeX formatting artifacts (27% vs. 64%). We establish the first benchmark for mathematical sentence recognition (S2L-sentences) and achieve an equation CER of 40%. This work lays the groundwork for future advances in multimodal AI, with a particular focus on mathematical content recognition.

cs.CV↗

arXiv2026

The Rogue Scalpel: Activation Steering Compromises LLM Safety

Activation steering is a promising technique for controlling LLM behavior by adding semantically meaningful vectors directly into a model's hidden states during inference. It is often framed as a precise, interpretable, and potentially safer alternative to fine-tuning. We demonstrate the opposite: steering systematically breaks model alignment safeguards, making it comply with harmful requests. Through extensive experiments on different model families, we show that even steering in a random direction can increase the probability of harmful compliance from 0% to 1-13%. Alarmingly, steering benign features from a sparse autoencoder (SAE), a common source of interpretable directions, demonstrates a comparable harmful potential. Finally, we show that combining 20 randomly sampled vectors that jailbreak a single prompt creates a universal attack, significantly increasing harmful compliance on unseen requests. These results challenge the paradigm of safety through interpretability, showing that precise control over model internals does not guarantee precise control over model behavior.

cs.LG↗

arXiv2025

HAMSA: Hijacking Aligned Compact Models via Stealthy Automation

Large Language Models (LLMs), especially their compact efficiency-oriented variants, remain susceptible to jailbreak attacks that can elicit harmful outputs despite extensive alignment efforts. Existing adversarial prompt generation techniques often rely on manual engineering or rudimentary obfuscation, producing low-quality or incoherent text that is easily flagged by perplexity-based filters. We present an automated red-teaming framework that evolves semantically meaningful and stealthy jailbreak prompts for aligned compact LLMs. The approach employs a multi-stage evolutionary search, where candidate prompts are iteratively refined using a population-based strategy augmented with temperature-controlled variability to balance exploration and coherence preservation. This enables the systematic discovery of prompts capable of bypassing alignment safeguards while maintaining natural language fluency. We evaluate our method on benchmarks in English (In-The-Wild Jailbreak Prompts on LLMs), and a newly curated Arabic one derived from In-The-Wild Jailbreak Prompts on LLMs and annotated by native Arabic linguists, enabling multilingual assessment.

cs.CL↗

arXiv2025

I Have Covered All the Bases Here: Interpreting Reasoning Features in Large Language Models via Sparse Autoencoders

Recent LLMs like DeepSeek-R1 have demonstrated state-of-the-art performance by integrating deep thinking and complex reasoning during generation. However, the internal mechanisms behind these reasoning processes remain unexplored. We observe reasoning LLMs consistently use vocabulary associated with human reasoning processes. We hypothesize these words correspond to specific reasoning moments within the models' internal mechanisms. To test this hypothesis, we employ Sparse Autoencoders (SAEs), a technique for sparse decomposition of neural network activations into human-interpretable features. We introduce ReasonScore, an automatic metric to identify active SAE features during these reasoning moments. We perform manual and automatic interpretation of the features detected by our metric, and find those with activation patterns matching uncertainty, exploratory thinking, and reflection. Through steering experiments, we demonstrate that amplifying these features increases performance on reasoning-intensive benchmarks (+2.2%) while producing longer reasoning traces (+20.5%). Using the model diffing technique, we provide evidence that these features are present only in models with reasoning capabilities. Our work provides the first step towards a mechanistic understanding of reasoning in LLMs. Code available at https://github.com/AIRI-Institute/SAE-Reasoning

cs.CL↗

arXiv2025

Geopolitical biases in LLMs: what are the "good" and the "bad" countries according to contemporary language models

This paper evaluates geopolitical biases in LLMs with respect to various countries though an analysis of their interpretation of historical events with conflicting national perspectives (USA, UK, USSR, and China). We introduce a novel dataset with neutral event descriptions and contrasting viewpoints from different countries. Our findings show significant geopolitical biases, with models favoring specific national narratives. Additionally, simple debiasing prompts had a limited effect in reducing these biases. Experiments with manipulated participant labels reveal models' sensitivity to attribution, sometimes amplifying biases or recognizing inconsistencies, especially with swapped labels. This work highlights national narrative biases in LLMs, challenges the effectiveness of simple debiasing methods, and offers a framework and dataset for future geopolitical bias research.

cs.CL↗

arXiv2025

CLEAR: Character Unlearning in Textual and Visual Modalities

Machine Unlearning (MU) is critical for removing private or hazardous information from deep learning models. While MU has advanced significantly in unimodal (text or vision) settings, multimodal unlearning (MMU) remains underexplored due to the lack of open benchmarks for evaluating cross-modal data removal. To address this gap, we introduce CLEAR, the first open-source benchmark designed specifically for MMU. CLEAR contains 200 fictitious individuals and 3,700 images linked with corresponding question-answer pairs, enabling a thorough evaluation across modalities. We conduct a comprehensive analysis of 11 MU methods (e.g., SCRUB, gradient ascent, DPO) across four evaluation sets, demonstrating that jointly unlearning both modalities outperforms single-modality approaches. The dataset is available at https://huggingface.co/datasets/therem/CLEAR

cs.CV↗

arXiv2025

Novel Loss-Enhanced Universal Adversarial Patches for Sustainable Speaker Privacy

Deep learning voice models are commonly used nowadays, but the safety processing of personal data, such as human identity and speech content, remains suspicious. To prevent malicious user identification, speaker anonymization methods were proposed. Current methods, particularly based on universal adversarial patch (UAP) applications, have drawbacks such as significant degradation of audio quality, decreased speech recognition quality, low transferability across different voice biometrics models, and performance dependence on the input audio length. To mitigate these drawbacks, in this work, we introduce and leverage the novel Exponential Total Variance (TV) loss function and provide experimental evidence that it positively affects UAP strength and imperceptibility. Moreover, we present a novel scalable UAP insertion procedure and demonstrate its uniformly high performance for various audio lengths.

cs.SD↗

arXiv2025

Optimal Hyperspectral Undersampling Strategy for Satellite Imaging

Hyperspectral image (HSI) classification presents significant challenges due to the high dimensionality, spectral redundancy, and limited labeled data typically available in real-world applications. To address these issues and optimize classification performance, we propose a novel band selection strategy known as Iterative Wavelet-based Gradient Sampling (IWGS). This method incrementally selects the most informative spectral bands by analyzing gradients within the wavelet-transformed domain, enabling efficient and targeted dimensionality reduction. Unlike traditional selection methods, IWGS leverages the multi-resolution properties of wavelets to better capture subtle spectral variations relevant for classification. The iterative nature of the approach ensures that redundant or noisy bands are systematically excluded while maximizing the retention of discriminative features. We conduct comprehensive experiments on two widely-used benchmark HSI datasets: Houston 2013 and Indian Pines. Results demonstrate that IWGS consistently outperforms state-of-the-art band selection and classification techniques in terms of both accuracy and computational efficiency. These improvements make our method especially suitable for deployment in edge devices or other resource-constrained environments, where memory and processing power are limited. In particular, IWGS achieved an overall accuracy up to 97.8% on Indian Pines for selected classes, confirming its effectiveness and generalizability across different HSI scenarios.

cs.CV↗

arXiv2025

FS-Net: Full Scale Network and Adaptive Threshold for Improving Extraction of Micro-Retinal Vessel Structures

Retinal vascular segmentation, a widely researched topic in biomedical image processing, aims to reduce the workload of ophthalmologists in treating and detecting retinal disorders. Segmenting retinal vessels presents unique challenges; previous techniques often failed to effectively segment branches and microvascular structures. Recent neural network approaches struggle to balance local and global properties and frequently miss tiny end vessels, hindering the achievement of desired results. To address these issues in retinal vessel segmentation, we propose a comprehensive micro-vessel extraction mechanism based on an encoder-decoder neural network architecture. This network includes residual, encoder booster, bottleneck enhancement, squeeze, and excitation building blocks. These components synergistically enhance feature extraction and improve the prediction accuracy of the segmentation map. Our solution has been evaluated using the DRIVE, CHASE-DB1, and STARE datasets, yielding competitive results compared to previous studies. The AUC and accuracy on the DRIVE dataset are 0.9884 and 0.9702, respectively. For the CHASE-DB1 dataset, these scores are 0.9903 and 0.9755, respectively, and for the STARE dataset, they are 0.9916 and 0.9750. Given its accurate and robust performance, the proposed approach is a solid candidate for being implemented in real-life diagnostic centers and aiding ophthalmologists.

eess.IV↗

arXiv2024

Certification of Speaker Recognition Models to Additive Perturbations

Speaker recognition technology is applied to various tasks, from personal virtual assistants to secure access systems. However, the robustness of these systems against adversarial attacks, particularly to additive perturbations, remains a significant challenge. In this paper, we pioneer applying robustness certification techniques to speaker recognition, initially developed for the image domain. Our work covers this gap by transferring and improving randomized smoothing certification techniques against norm-bounded additive perturbations for classification and few-shot learning tasks to speaker recognition. We demonstrate the effectiveness of these methods on VoxCeleb 1 and 2 datasets for several models. We expect this work to improve the robustness of voice biometrics and accelerate the research of certification methods in the audio domain.

cs.SD↗

arXiv2024

GLiRA: Black-Box Membership Inference Attack via Knowledge Distillation

While Deep Neural Networks (DNNs) have demonstrated remarkable performance in tasks related to perception and control, there are still several unresolved concerns regarding the privacy of their training data, particularly in the context of vulnerability to Membership Inference Attacks (MIAs). In this paper, we explore a connection between the susceptibility to membership inference attacks and the vulnerability to distillation-based functionality stealing attacks. In particular, we propose {GLiRA}, a distillation-guided approach to membership inference attack on the black-box neural network. We observe that the knowledge distillation significantly improves the efficiency of likelihood ratio of membership inference attack, especially in the black-box setting, i.e., when the architecture of the target model is unknown to the attacker. We evaluate the proposed method across multiple image classification datasets and models and demonstrate that likelihood ratio attacks when guided by the knowledge distillation, outperform the current state-of-the-art membership inference attacks in the black-box setting.

cs.LG↗