Hugging Face Transformers Regular Expression Denial of Service (ReDoS) vulnerability

A Regular Expression Denial of Service ReDoS vulnerability exists in the Hugging Face Transformers library, specifically in the converttfweightnametoptweightname function. This function, responsible for converting TensorFlow weight names to PyTorch format, uses a regex pattern /^/^// that can be...

CVE-2025-5197

The CVE-2025-5197 ReDoS vulnerability affects Hugging Face Transformers in the convert_tf_weight_name_to_pt_weight_name() function, where the regex /[^/]___([^/] )/ can cause excessive CPU usage via catastrophic backtracking. Affected versions: up to 4.51.3, with a fix in 4.53.0. Practical impact...

PT-2025-32158

Name of the Vulnerable Software and Affected Versions Hugging Face Transformers versions up to 4.51.3 Description A Regular Expression Denial of Service ReDoS vulnerability exists in the convert tf weight name to pt weight name function of the Hugging Face Transformers library. This function, whi...

CompLeak: Deep Learning Model Compression Exacerbates Privacy Leakage

Model compression is crucial for minimizing memory storage and accelerating inference in deep learning DL models, including recent foundation models like large language models LLMs. Users can access different compressed model versions according to their resources and budget. However, while existi...

Multi-Trigger Poisoning Amplifies Backdoor Vulnerabilities in LLMs

Recent studies have shown that Large Language Models LLMs are vulnerable to data poisoning attacks, where malicious training examples embed hidden behaviours triggered by specific input patterns. However, most existing works assume a phrase and focus on the attack's effectiveness, offering limite...

Generalized and Unified Equivalences between Hardness and Pseudoentropy

Pseudoentropy characterizations provide a quantitatively precise demonstration of the close relationship between computational hardness and computational randomness. We prove a unified pseudoentropy characterization that generalizes and strengthens previous results for both uniform and non-unifor...

Cyclic Equalizability of Words and Its Application to Card-Based Cryptography

Card-based cryptography is a research area to implement cryptographic procedures using a deck of physical cards. In recent years, it has been found to be related to finite group theory and algebraic combinatorics, and is becoming more and more closely connected to the field of mathematics. In thi...

Towards Provable (In)Secure Model Weight Release Schemes

Recent secure weight release schemes claim to enable open-source model distribution while protecting model ownership and preventing misuse. However, these approaches lack rigorous security foundations and provide only informal security guarantees. Inspired by established works in cryptography, we...

New Characterization of Full Weight Spectrum One-Orbit Cyclic Subspace Codes

In this paper, we determine the weight distributions of a family of FWS codes and exhibit some equivalence classes of FWS codes under certain conditions. Furthermore, we provide a complete classification for $r$-FWS codes...

SecureFed: a Two-Phase Framework for Detecting Malicious Clients in Federated Learning

Federated Learning FL protects data privacy while providing a decentralized method for training models. However, because of the distributed schema, it is susceptible to adversarial clients that could alter results or sabotage model performance. This study presents SecureFed, a two-phase FL...

Regular Expression Denial of Service (ReDoS) in AdamWeightDecay Optimizer

The AdamWeightDecay optimizer is vulnerable to Regular Expression Denial of Service ReDoS. If an attacker can control the patterns in the includeinweightdecay or excludefromweightdecay lists, they can provide a malicious regular expression that causes catastrophic backtracking. When the optimizer...

PT-2025-39174

Name of the Vulnerable Software and Affected Versions huggingface/transformers versions prior to 4.53.0 Description The software is susceptible to a Regular Expression Denial of Service ReDoS within the AdamWeightDecay optimizer. The issue stems from the do use weight decay method, which handles...

Astra Linux – Vulnerability found in Linux 6.1, Linux 6.12

In the Linux kernel, the following vulnerability has been resolved: net: let netcore.devweight always be non-zero The following problem was encountered during stability testing: NULL netdevice: The NAPI poll function processbacklog+0x0/0x530 returned 1, exceeding its budget of 0. ------------ Cut...

MixBridge: Heterogeneous Image-To-Image Backdoor Attack through Mixture of Schrödinger Bridges

This paper focuses on implanting multiple heterogeneous backdoor triggers in bridge-based diffusion models designed for complex and arbitrary input distributions. Existing backdoor formulations mainly address single-attack scenarios and are limited to Gaussian noise input models. To fill this gap...

CVE-2023-37388

Auth. admin+ Stored Cross-Site Scripting XSS vulnerability in Sudipto Pratap Mahato Simple Light Weight Social Share plugin = 2.0 versions...

CVE-2022-46794

Cross-Site Request Forgery CSRF vulnerability in weightbasedshipping.Com WooCommerce Weight Based Shipping plugin = 5.4.1 versions...

CVE-2022-29791

The HiAIserver has a vulnerability in verifying the validity of the weight used in the model.Successful exploitation of this vulnerability will affect AI services...

CVE-2022-29796

The HiAIserver has a vulnerability in verifying the validity of the weight used in the model.Successful exploitation of this vulnerability will affect AI services...

CVE-2022-22261

The HiAIserver has a vulnerability in verifying the validity of the weight used in the model.Successful exploitation of this vulnerability will affect AI services...

CVE-2022-36027

TensorFlow is an open source platform for machine learning. When converting transposed convolutions using per-channel weight quantization the converter segfaults and crashes the Python process. We have patched the issue in GitHub commit aa0b852a4588cea4d36b74feb05d93055540b450. The fix will be...