Searcharxiv⌕ Search
Home/Archive/arXiv

arXiv · 2610.04378

COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol Layers

Nahom Birhan·Mehrdad Rostamzadeh·Sidhant Narula·Mahmoud Nazzal·Mohammad Ghasemigol·Daniel Takabi
View original source ↗Read source document ↗Export metadata ↓

Abstract

Large language models increasingly mediate tool use in Model Context Protocol (MCP) systems, where adversarial influence may enter through user instructions, tool schemas, tool outputs, or protocol messages. Existing benchmarks often evaluate deployed agents, conflating model susceptibility with guardrails, orchestration, and general task capability. We introduce COPEX (COntext Provider EXploitation), a controlled benchmark that isolates the model as an MCP client by fixing the surrounding agent stack and varying only the tool-selecting model. COPEX covers 25 attack types instantiated as 125 scenarios across four entry surfaces: model/agent, client, server/tool, and transport. Across nine models and 3,375 trials, the mean attack success rate is 64.4%, with surface-level means ranging from 58.3% to 71.4%. Some client- and transport-level attacks succeed partly outside the model's observation or control, separating system exposure from model susceptibility. Combined input and context scanning reduces mean attack success by 49.6% on an eight-attack defense subset relative to the undefended setting. The benchmark is available at https://github.com/inspire-center/copex.

Explore related subjects

cs.CRcs.AI

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Nahom Birhan, Mehrdad Rostamzadeh, Sidhant Narula, Mahmoud Nazzal, Mohammad Ghasemigol, Daniel Takabi. 2026-10-03. COPEX: Benchmarking LLM Robustness to Adversarial Context Across Model Context Protocol Layers. https://arxiv.org/abs/2610.04378

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Large Language Models for Cryptocurrency Transaction Analysis: A Bitcoin Case Study

Cryptocurrencies are widely used, yet current methods for analyzing transactions often rely on opaque, black-box models. While these models may achieve high performance, their outputs are usually difficult to interpret and adapt, making it challenging to capture nuanced behavioral patterns. Large language models (LLMs) have the potential to address these gaps, but their capabilities in this area remain largely unexplored, particularly in cybercrime detection. In this paper, we test this hypothesis by applying LLMs to real-world cryptocurrency transaction graphs, with a focus on Bitcoin, one of the most studied and widely adopted blockchain networks. We introduce a three-tiered framework to assess LLM capabilities: foundational metrics, characteristic overview, and contextual interpretation. This includes a new, human-readable graph representation format, LLM4TG, and a connectivity-enhanced transaction graph sampling algorithm, CETraS. Together, they significantly reduce token requirements, transforming the analysis of multiple moderately large-scale transaction graphs with LLMs from nearly impossible to feasible under strict token limits. Experimental results demonstrate that LLMs have outstanding performance on foundational metrics and characteristic overview, where the accuracy of recognizing most basic information at the node level exceeds 98.50% and the proportion of obtaining meaningful characteristics reaches 95.00%. Regarding contextual interpretation, LLMs also demonstrate strong performance in classification tasks, even with very limited labeled data, where top-3 accuracy reaches 72.43% with explanations. While the explanations are not always fully accurate, they highlight the strong potential of LLMs in this domain. At the same time, several limitations persist, which we discuss along with directions for future research.

cs.CR↗
arXiv2026

Understanding the Identity-Transformation Approach in OIDC-Compatible Privacy-Preserving SSO Services

Single sign-on (SSO) enables a user to log into multiple websites, called relying parties (RPs), by her username and credential set up in another trusted web system, called the identity provider (IdP). Identity transformations are proposed in UppreSSO to provide privacy-preserving SSO services, preventing both IdP-based login tracing and RP-based identity linkage. While the security and privacy guarantees of UppreSSO have been proved, several essential issues on the identity-transformation approach are not well studied. In this paper, we comprehensively investigate this approach as below. Firstly, several suggestions to efficiently integrate identity transformations into OpenID Connect (OIDC) are explained. Then, we uncover the relationship between identity transformations in SSO and oblivious pseudo-random functions (OPRFs), and present two variations of the properties required for SSO security as well as other requirements, to analyze existing OPRF protocols. Finally, new identity transformations different from those proposed in UppreSSO, are constructed based on some OPRFs. To the best of our knowledge, this is the first time to uncover the relationship between identity transformations in SSO services and OPRFs, and prove the SSO-related properties (i.e., output uniqueness, key-identifier freeness, and collision resistance on 1st/2nd-input) of typical OPRFs.

cs.CR↗
arXiv2026

Practical Feasibility of Gradient Inversion Attacks in Federated Learning

Gradient inversion attacks are often presented as a serious privacy threat in federated learning, with recent work reporting increasingly strong reconstructions under favorable experimental settings. However, it remains unclear whether such attacks are feasible in modern, performance-optimized systems deployed in practice. In this work, we evaluate the practical feasibility of gradient inversion for image-based federated learning. We conduct a systematic study across multiple datasets and tasks, including image classification and object detection, using canonical vision architectures at contemporary resolutions. Our results show that while gradient inversion remains possible for certain legacy or transitional designs under highly restrictive assumptions, modern, performance-optimized models consistently resist meaningful reconstruction visually. We further demonstrate that many reported successes rely on upper-bound settings, such as inference mode operation or architectural simplifications which do not reflect realistic training pipelines. Taken together, our findings indicate that, under an honest-but-curious server assumption, high-fidelity image reconstruction via gradient inversion does not constitute a critical privacy risk in production-optimized federated learning systems, and that practical risk assessments must carefully distinguish diagnostic attack settings from real-world deployments.

cs.CR↗