LLM Unlearning Should Be Form-Independent

Large Language Model LLM unlearning aims to erase or suppress undesirable knowledge within the model, offering promise for controlling harmful or private information to prevent misuse. However, recent studies highlight its limited efficacy in real-world scenarios, hindering practical adoption. In...

FedShield-LLM: a Secure and Scalable Federated Fine-Tuned Large Language Model

Federated Learning FL offers a decentralized framework for training and fine-tuning Large Language Models LLMs by leveraging computational resources across organizations while keeping sensitive data on local devices. It addresses privacy and security concerns while navigating challenges associate...

Learning to Diagnose Privately: DP-Powered LLMs for Radiology Report Classification

Purpose: This study proposes a framework for fine-tuning large language models LLMs with differential privacy DP to perform multi-abnormality classification on radiology report text. By injecting calibrated noise during fine-tuning, the framework seeks to mitigate the privacy risks associated wit...

SHE-LoRA: Selective Homomorphic Encryption for Federated Tuning with Heterogeneous LoRA

Federated fine-tuning of large language models LLMs is critical for improving their performance in handling domain-specific tasks. However, prior work has shown that clients' private data can actually be recovered via gradient inversion attacks. Existing privacy preservation techniques against su...

CVE-2024-3665

The Rank Math SEO with AI SEO Tools plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the plugin's HowTo and FAQ widgets in all versions up to, and including, 1.0.216 due to insufficient input sanitization and output escaping on user supplied attributes. This makes it possible...

CVE-2024-2536

The Rank Math SEO with AI SEO Tools plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the HowTo block attributes in all versions up to, and including, 1.0.214 due to insufficient input sanitization and output escaping on user supplied attributes. This makes it possible for...

CVE-2024-4627

The Rank Math SEO WordPress plugin before 1.0.219 does not sanitise and escape some of its settings, which could allow users with access to the General Settings by default admin, however such access can be given to lower roles via the Role Manager feature of the Rank Math SEO WordPress plugin...

CVE-2024-4617

The Rank Math SEO with AI Best SEO Tools plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the ‘id’ parameter in versions up to, and including, 1.0.218 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with...

CVE-2024-4335

The Rank Math SEO with AI Best SEO Tools plugin for WordPress is vulnerable to Stored Cross-Site Scripting via the ‘textAlign’ parameter in versions up to, and including, 1.0.217 due to insufficient input sanitization and output escaping. This makes it possible for authenticated attackers, with...

CVE-2024-9161

The Rank Math SEO – AI SEO Tools to Dominate SEO Rankings plugin for WordPress is vulnerable to unauthorized modification and loss of data due to a missing capability check on the 'updatemetadata' function in all versions up to, and including, 1.0.228. This makes it possible for unauthenticated...

CVE-2023-38675

FPE in paddle.linalg.matrixrank in PaddlePaddle before 2.6.0. This flaw can cause a runtime crash and a denial of service...

CVE-2023-32600

Auth. contributor+ Stored Cross-Site Scripting XSS vulnerability in Rank Math SEO plugin = 1.0.119 versions...

CVE-2023-33976

TensorFlow is an end-to-end open source platform for machine learning. arrayops.upperbound causes a segfault when not given a rank 2 tensor. The fix will be included in TensorFlow 2.13 and will also cherrypick this commit on TensorFlow 2.12...

CVE-2023-27707

SQL injection vulnerability found in DedeCMS v.5.7.106 allows a remote attacker to execute arbitrary code via the rank parameter in the /dede/groupstore.php endpoint...

CVE-2023-23888

Improper Limitation of a Pathname to a Restricted Directory 'Path Traversal' vulnerability in Rank Math Rank Math SEO allows Path Traversal.This issue affects Rank Math SEO: from n/a through 1.0.107.2...

CVE-2023-32800

Unauth. Reflected Cross-Site Scripting XSS vulnerability in One Rank Math SEO PRO plugin = 3.0.35 versions...

CVE-2022-36019

TensorFlow is an open source platform for machine learning. If FakeQuantWithMinMaxVarsPerChannel is given min or max tensors of a rank other than one, it results in a CHECK fail that can be used to trigger a denial of service attack. We have patched the issue in GitHub commit...

CVE-2022-41899

TensorFlow is an open source platform for machine learning. Inputs densefeatures or examplestatedata not of rank 2 will trigger a CHECK fail in SdcaOptimizer. We have patched the issue in GitHub commit 80ff197d03db2a70c6a111f97dcdacad1b0babfa. The fix will be included in TensorFlow 2.11. We will...

CVE-2022-35988

TensorFlow is an open source platform for machine learning. When tf.linalg.matrixrank receives an empty input a, the GPU kernel gives a CHECK fail that can be used to trigger a denial of service attack. We have patched the issue in GitHub commit c55b476aa0e0bd4ee99d0f3ad18d9d706cd1260a. The fix...

CVE-2021-29544

TensorFlow is an end-to-end open source platform for machine learning. An attacker can trigger a denial of service via a CHECK-fail in tf.rawops.QuantizeAndDequantizeV4Grad. This is because the implementation does not validate the rank of the input tensors. In turn, this results in the tensors...