Awakening the Hydra: Stabilizing Multi-Concept Backdoor Injection in Text-To-Image Diffusion Models

Text-to-image diffusion models are increasingly developed through open-source reuse and repeated downstream fine-tuning, where reused checkpoints are difficult to verify and thus more susceptible to hidden backdoor behaviors. In such ecosystems, a single pretrained model may be sequentially adapt...

An ICE Firearms Trainer Was Involved in At Least 4 Deadly Shootings

David Norman, a former Phoenix police officer who’s described himself as “a fucking savage,” now runs a company that provided training to Homeland Security’s Special Response Teams...

Docker_MCP_POC

Java vulnerable scan POC Minimal Maven project used to comp...

A No-Defense Defense against Gradient-Based Adversarial Attacks on ML-NIDS: Is Less More?

Gradient-based adversarial attacks subtly manipulate inputs of Machine Learning ML models to induce incorrect predictions. This paper investigates whether careful architectural choices alone can yield an inherently robust Deep Neural Network DNN-based Network Intrusion Detection Systems NIDS,...

Backdooring Masked Diffusion Language Models

Masked diffusion language models MDLMs are emerging as a compelling new paradigm for text generation, but their training-time security remains largely unexplored. Existing backdoor attacks on Gaussian diffusion models or autoregressive language models do not directly apply to MDLMs because MDLMs...

Exploit for CVE-2026-6857

⚠️ Security Research & Legal Disclaimer 📌 Purpose of This...

a2 (>=0.1.0 <=0.3.17), abnativ (>=1.1.0 <=1.2.9) +356 more potentially affected by CVE-2026-2652 via mlflow (>=0.8.2 <=3.10.1)

mlflow PYPI version =0.8.2, =0.1.0, =1.1.0, =0.0.5, =0.1.0, =0.1.0, =1.7.0, =1.7.0, =1.8.0, =1.7.0, =1.7.0, =0.1.1, =0.1.5 - anovos =1.1.0 and more Source cves: CVE-2026-2652 Source advisory: OSV:GHSA-75CM-X2W3-8MGF...

Access Control Bypass

Overview Affected versions of this package are vulnerable to Access Control Bypass via the logs and stats API actions in the RoutineViewSet process. An attacker can access another user's private workout session notes, exercise history, and training statistics by enumerating public template routin...

Exploit for CVE-2026-46300

⚠️ Security Research & Legal Disclaimer 📌 Purpose of This...

WARD: Adversarially Robust Defense of Web Agents against Prompt Injections

Web agents can autonomously complete online tasks by interacting with websites, but their exposure to open web environments makes them vulnerable to prompt injection attacks embedded in HTML content or visual interfaces. Existing guard models still suffer from limited generalization to unseen...

PT-2026-41136

Name of the Vulnerable Software and Affected Versions wger versions prior to 2.5.0a2 Description An insecure direct object reference IDOR allows any authenticated user to access the private health data of other users. The issue occurs when a user has a routine marked as a public template, which...

CVE-2026-43477

In the Linux kernel, the following vulnerability has been resolved: drm/i915/vrr: Configure VRR timings after enabling TRANSDDIFUNCCTL Apparently ICL may hang with an MCE if we write TRANSVRRVMAX/FLIPLINE before enabling TRANSDDIFUNCCTL. Personally I was only able to reproduce a hang on an Dell X...

CVE-2026-43477 drm/i915/vrr: Configure VRR timings after enabling TRANS_DDI_FUNC_CTL

In the Linux kernel, the following vulnerability has been resolved: drm/i915/vrr: Configure VRR timings after enabling TRANSDDIFUNCCTL Apparently ICL may hang with an MCE if we write TRANSVRRVMAX/FLIPLINE before enabling TRANSDDIFUNCCTL. Personally I was only able to reproduce a hang on an Dell X...

Code-Centric Detection of Vulnerability-Fixing Commits: A Unified Benchmark and Empirical Study

Automated detection of vulnerability-fixing commits VFCs is critical for timely security patch deployment, as advisory databases lag patch releases by a median of 25 days and many fixes never receive advisories. We present a comprehensive evaluation of code language model based VFC detection...

CVE-2026-31254

The flash-attention project thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 2025-13-04 contains a code injection vulnerability CWE-94 in its training script. The script registers the Python eval function as a Hydra configuration resolver under the name eval. This allows configuration files t...

CVE-2026-31218

The loadmodel function in the neuralmagictraining.py script of the optimate project in commit a6d302f912b481c94370811af6b11402f51d377f 2024-07-21 is vulnerable to insecure deserialization CWE-502. When loading a model state dictionary from a statedict.pt file via torch.load, the function does not...

CVE-2026-31218

The loadmodel function in the neuralmagictraining.py script of the optimate project in commit a6d302f912b481c94370811af6b11402f51d377f 2024-07-21 is vulnerable to insecure deserialization CWE-502. When loading a model state dictionary from a statedict.pt file via torch.load, the function does not...

CVE-2026-31217

The loadmodel function in the neuralmagictraining.py script of the optimate project in commit a6d302f912b481c94370811af6b11402f51d377f 2024-07-21 allows arbitrary code execution. When a user supplies a directory path via the --model command-line argument, the function reads a module.py file from...

a2 (>=0.1.0 <=0.3.17), abnativ (>=1.1.0 <=1.2.9) +346 more potentially affected by CVE-2026-2614 via mlflow (>=0.8.2 <=3.0.1)

mlflow PYPI version =0.8.2, =0.1.0, =1.1.0, =0.0.5, =0.1.0, =0.1.0, =1.7.0, =1.7.0, =1.8.0, =1.7.0, =1.7.0, =0.1.1, =0.1.5 - anovos =1.1.0 and more Source cves: CVE-2026-2614 Source advisory: OSV:GHSA-42H5-H8QH-VV9V...

EUVD-2026-29101

The flash-attention project thru commit e724e2588cbe754beb97cf7c011b5e7e34119e62 2025-13-04 contains a code injection vulnerability CWE-94 in its training script. The script registers the Python eval function as a Hydra configuration resolver under the name eval. This allows configuration files t...