1413 matches found
Moderate: Red Hat Security Advisory: Red Hat AI Inference Server 3.0 (CUDA)
Red Hat AI Inference Server 3.0 CUDA is now available. Red Hat® AI Inference Server...
Moderate: Red Hat Security Advisory: Red Hat AI Inference Server 3.0 (ROCm)
Red Hat AI Inference Server 3.0 ROCm is now available. Red Hat® AI Inference Server...
Cascade: Token-Sharded Private LLM Inference
As LLMs continue to increase in parameter size, the computational resources required to run them are available to fewer parties. Therefore, third-party inference services -- where LLMs are hosted by third parties with significant computational resources -- are becoming increasingly popular...
CVE-2025-6920
A flaw was found in the authentication enforcement mechanism of a model inference API in ai-inference-server. All /v1/ endpoints are expected to enforce API key validation. However, the POST /invocations endpoint failed to do so, resulting in an authentication bypass. This vulnerability allows...
CVE-2025-6920 Ai-inference-server: authentication bypass via unprotected inference endpoint in api
A flaw was found in the authentication enforcement mechanism of a model inference API in ai-inference-server. All /v1/ endpoints are expected to enforce API key validation. However, the POST /invocations endpoint failed to do so, resulting in an authentication bypass. This vulnerability allows...
CVE-2025-6920 Ai-inference-server: authentication bypass via unprotected inference endpoint in api
A flaw was found in the authentication enforcement mechanism of a model inference API in ai-inference-server. All /v1/ endpoints are expected to enforce API key validation. However, the POST /invocations endpoint failed to do so, resulting in an authentication bypass. This vulnerability allows...
CVE-2025-6920
CVE-2025-6920 affects ai-inference-server: the POST /invocations endpoint bypasses API key validation, permitting unauthorized access to inference features and potentially backend resources. Affected: model inference API under /v1/*; root cause: authentication enforcement failure on /invocations....
PT-2025-27527 · Unknown · Ai-Inference-Server
Name of the Vulnerable Software and Affected Versions: ai-inference-server affected versions not specified Description: A flaw was found in the authentication enforcement mechanism of a model inference API. The issue affects the "/v1/" endpoints, where API key validation is expected but not...
CVE-2025-52566
llama.cpp is an inference of several LLM models in C/C++. Prior to version b5721, there is a signed vs. unsigned integer overflow in llama.cpp's tokenizer implementation llamavocab::tokenize src/llama-vocab.cpp:3036 resulting in unintended behavior in tokens copying size comparison. Allowing...
Amplifying Machine Learning Attacks through Strategic Compositions
Machine learning ML models are proving to be vulnerable to a variety of attacks that allow the adversary to learn sensitive information, cause mispredictions, and more. While these attacks have been extensively studied, current research predominantly focuses on analyzing each attack type...
Image Corruption-Inspired Membership Inference Attacks against Large Vision-Language Models
Large vision-language models LVLMs have demonstrated outstanding performance in many downstream tasks. However, LVLMs are trained on large-scale datasets, which can pose privacy risks if training images contain sensitive information. Therefore, it is important to detect whether an image is used t...
SecONNds: Secure Outsourced Neural Network Inference on ImageNet
The widespread adoption of outsourced neural network inference presents significant privacy challenges, as sensitive user data is processed on untrusted remote servers. Secure inference offers a privacy-preserving solution, but existing frameworks suffer from high computational overhead and...
HE-LRM: Encrypted Deep Learning Recommendation Models Using Fully Homomorphic Encryption
Fully Homomorphic Encryption FHE is an encryption scheme that not only encrypts data but also allows for computations to be applied directly on the encrypted data. While computationally expensive, FHE can enable privacy-preserving neural inference in the client-server setting: a client encrypts...
UCD: Unlearning in LLMs Via Contrastive Decoding
Machine unlearning aims to remove specific information, e.g. sensitive or undesirable content, from large language models LLMs while preserving overall performance. We propose an inference-time unlearning algorithm that uses contrastive decoding, leveraging two auxiliary smaller models, one train...
Rectifying Privacy and Efficacy Measurements in Machine Unlearning: a New Inference Attack Perspective
Machine unlearning focuses on efficiently removing specific data from trained models, addressing privacy and compliance concerns with reasonable costs. Although exact unlearning ensures complete data removal equivalent to retraining, it is impractical for large-scale models, leading to growing...
Don't Throw the Baby out with the Bathwater: How and Why Deep Learning for ARC
The Abstraction and Reasoning Corpus ARC-AGI presents a formidable challenge for AI systems. Despite the typically low performance on ARC, the deep learning paradigm remains the most effective known strategy for generating skillful state-of-the-art neural networks NN across varied modalities and...
Enhancing One-run Privacy Auditing with Quantile Regression-Based Membership Inference
Differential privacy DP auditing aims to provide empirical lower bounds on the privacy guarantees of DP mechanisms like DP-SGD. While some existing techniques require many training runs that are prohibitively costly, recent work introduces one-run auditing approaches that effectively audit DP-SGD...
ReDASH: Fast and efficient Scaling in Arithmetic Garbled Circuits for Secure Outsourced Inference
Whitepaper called ReDASH: Fast and efficient Scaling in Arithmetic Garbled Circuits for Secure Outsourced Inference...
Black-Box Privacy Attacks on Shared Representations in Multitask Learning
Multitask learning MTL has emerged as a powerful paradigm that leverages similarities among multiple learning tasks, each with insufficient samples to train a standalone model, to solve them simultaneously while minimizing data sharing across users and organizations. MTL typically accomplishes th...