gradient-untangler

gradient-untangler A local research harness that searches for the exact tokens that make an open-weight language model start its answer the way you specify. Open-weight models ship as ordinary files: config.json, a tokenizer, and one or more .safetensors or .bin shards. A Hugging Face id is only ...

MIG-COW: Transferable Adversarial Attacks on Deepfake Detectors Via Gradient Decomposition

Despite recent advances, deepfake detectors remain vulnerable to adversarial examples, particularly in diverse, real-world settings. We propose MIG-COW, a novel adversarial attack framework that generates highly generalizable and visually imperceptible adversarial examples. By combining...

On the Adversarial Vulnerability of Neural Receivers with Unrolled LDPC Decoders

Neural receivers for 5G New Radio NR match classical model-based receivers in link-level performance, but their robustness to adversarial perturbations at the receiver input remains underexplored. We show that standard gradient-based attacks applied end-to-end through an unrolled low-density...

Empirical Evaluation of Machine Learning for Cloud Cryptojacking Detection: Effects of Preprocessing and Evaluation Protocols

Artificial intelligence AI has been widely applied to cloud cryptojacking detection. Classical machine learning models remain among the most widely used approaches for cloud cryptojacking detection. This paper presents an empirical evaluation of six widely used classical machine learning models f...

Mitigating Black-Box Score-Based Query Attacks Via Gradient Confusion

Score-based black-box adversarial attacks pose a serious threat to deployed machine learning systems by exploiting repeated model queries to craft adversarial examples, requiring neither access to model parameters nor architecture. Existing defenses predominantly rely on stochastic output...

Removing Information Content Does Not Certify Tamper Resistance in Open-Weight Models

Does removing harmful information make open-weight models resistant to fine-tuning attacks? We show that mutual information at release alone cannot universally certify slow recovery. Function-preserving reparameterizations leave information unchanged while altering gradient-descent geometry, so a...

Red-TTT: Test-Time Training for Automated Jailbreaking Large Language Models

Large language models remain vulnerable to jailbreaks, and automated red teaming is the standard way to find jailbreaks in large language models at scale. Current methods either draw more samples at test time through search, rewriting, and tree expansion, or train a stronger attacker offline with...

Does the Readout Bypass Leak the Input? A Feature-Visibility Audit of Hybrid Quantum-Classical Models

Readout-side residual hybrids concatenate raw inputs with measured quantum features. Under single-example gradient sharing, a biased first linear layer admits standard analytic recovery of its input, so the bypass exposes raw coordinates without requiring inversion of the quantum circuit. We audi...

Does the Unsafe Gradient Survive a Conversation? on the Fragility of Gradient-Based Jailbreak Detection in Multi-Turn Dialogue

Safety-aligned language models are commonly deployed as multi-turn assistants, which lets adversaries spread unsafe intent across several user turns instead of a single prompt. Gradient-based jailbreak detectors such as GradSafe were developed for single prompts: they score an input by the...

Forensic-Aware Continual Adaptation for Image Forgery Localization

The rapid evolution of image manipulation techniques has raised growing public security concerns. Existing Image Forgery Localization IFL methods can accurately localize manipulated regions but are often unable to adapt to newly emerging forgeries. In real-world forensic scenarios, data typically...

Inference-Layer Security: Defending against Adversarial Inference and Infrastructure Abuse

A Technical Report: Operating a large language model LLM as a service requires more than inference infrastructure: the provider must also defend against adversarial interactions that seek to exploit the service, including jailbreaking for harmful use, sophisticated denial of service, and...

Decoding Guardrails: XAI-Guided Perturbation Analysis of Prompt Injection Detection

Large language models LLMs are increasingly deployed in production systems, raising concerns about their exposure to adversarial manipulation through prompt injection and jailbreak attacks. Classifier-based guardrails, such as Prompt Guard 2, are widely used as a first line of defense against suc...

Towards TEE-Certified DP: Verifiable Differentially Private Training on Legacy GPUs

Wide adoption of machine learning has created growing policy and regulatory demand for protecting sensitive training data, with differential privacy DP emerging as a key mechanism. Yet a less-studied problem is how to certify the faithful execution of DP during training: an external verifier shou...

Astra Linux – Vulnerability in libvncserver

LibVNCClient is a library for easy implementation of a VNC client. In versions 0.9.15 and earlier, LibVNCClient’s Tight encoding decoder used fixed-size 2048-pixel scratch buffers for the Gradient filter. However, it did not reject Tight rectangles whose width exceeded 2048 pixels. A malicious VN...

Cascading Gradient Inversion Via LT-Code Inspired Peeling in Federated Learning

Federated learning shares model updates rather than raw data, yet these updates can be inverted to reconstruct the clients' training data. Analytic reconstruction attacks, which invert a gradient in closed form, degrade as the batch grows: prior single-round attacks recover only about half of a...

Federated Attack Campaign Detection Via Contrastive Encoding of Threat Indicators in Gradient Updates

Detecting orchestrated cyberattack campaigns that span multiple organizations traditionally requires sharing sensitive telemetry and threat intelligence across institutional boundaries and country borders, a barrier that Federated Learning removes by training shared threat detectors directly on...

Malicious code in arcus-gradient-colors (npm)

--- Credit: OpenSSF source...

Malicious code in arcus-gradient (npm)

--- Credit: OpenSSF source...

MGASA-2026-0315 Updated libvncserver packages fix security vulnerabilities

The updated packages fix security vulnerabilities: Heap Out-of-Bounds Read in HandleUltraZipBPP due to unchecked subrectangle count. CVE-2026-32853 NULL pointer dereferences in httpd proxy handlers via malformed CONNECT/GET requests. CVE-2026-32854 LibVNCClient Tight Gradient decoding allows...

Malicious code in better-gradient (PyPI)

Source: kam193 61ee3d2a1011f83d233eed4719b397bc9a4e69c449841335b187b323d9f980fc --- Category: MALICIOUS - The campaign has clearly malicious intent, like infostealers. Campaign: duvet-love-odyssey Reasons based on the campaign: - infostealer --- Credit: OpenSSF source...