CORTEXA
← Browse
arxivcs.CRcs.AI2026-07-21

Integrity of peer-to-peer distributed LLM inference under malicious nodes

Mert Cihangiroglu, Antonino Nocera

Peer-to-peer distributed inference executes a Large Language Model (LLM) on pooled consumer hardware by spreading its layers across many nodes. Every request passes through nodes that are owned and controlled by multiple independent parties. However, in this setting, any party can tamper with the output of its layers to corrupt the end result. Recomputing the forward pass on trusted hardware can catch this, but it introduces additional computational cost. The scientific literature includes several prior integrity-checking approaches, such as known-answer traps for image classifiers and cryptographic commitments. However, these solutions test only the exact correctness and do not account for the ordinary variation that may arise between benign nodes. In this paper, we propose a method that checks the output integrity by measuring the variation in the activations that each node passes to the next. A peer who wants to use the network selects a small set of secret canary inputs whose correct activations are known in advance and mixes them into regular traffic. Because the peers cannot tell a canary from a real query, any tampering node corrupts them as well. The deviation from the known reference then reveals malicious activity: benign nodes exhibit only minor variation from hardware-induced noise, whereas tampered nodes deviate far more. We treat the identification of malicious nodes as a probabilistic test that separates two drift distributions, without relying on a fixed threshold. We study 408 configurations with metrics and success criteria fixed before any experiment ran; the detector reaches AUROC 1.0, correctly ranking the malicious shard above every benign shard on every canary in every configuration.

View free PDFSource page

Related papers

arxivcs.CRcs.AI2026-07-12

PromptGraph: Graph-Guided Prompt Sanitization for Balancing Privacy and Utility in LLM Inference

Chen Gu, Hui Wan, Donghui Hu, Hui Wang, Zhuoer Gu

Large Language Model (LLM) services introduce a fundamental privacy challenge. Sensitive information may be inferred not only from explicit identifiers, such as names or phone numbers, but also from contextual associations among otherwise innocuous spans. Existing sanitizers typi…

View free PDFSource page
arxivcs.LGcs.AIcs.CLcs.CR2026-07-01

Black-Box Inference of LLM Architectural Properties with Restrictive API Access

Christopher Ellis, Shreyas Chaudhari, Mei-Yu Wang, Leighton Barnes, Giulia Fanti, José M. F. Moura

In practice, most commercial LLM providers do not publicly release details of underlying LLM architectures. However, prior work has shown that given limited API access to an LLM (namely, top-$k$ logits and/or a logit bias function), one can recover certain architectural details o…

View free PDFSource page
arxivcs.CRcs.AIcs.SE2026-07-03

LLM-Enhanced Hierarchical Heterogeneous Graph Representation Learning for Malicious Python Package Detection

Hang Gao, Xiaoyu Chen, Baoquan Cui, Zhen Tang, Peng Qiao, Fengge Wu, et al.

Malicious Python packages have become a major threat to software supply chain ecosystems due to the widespread adoption of open-source repositories such as PyPI. Existing learning-based detection methods struggle to capture the hierarchical organization and heterogeneous interact…

View free PDFSource page
arxivcs.CRcs.AI2026-07-14

Efficient and Privacy Aware Edge Cloud Collaborative Inference for Large Language Models

Cheng Li, Jiexiong Liu, Yixuan Chen, Yi Li

On-device LLM inference faces a trilemma of response latency, limited hardware resources and user privacy. Full cloud inference delivers strong computing power but exposes user prompts and dialogue data, while standalone on-device inference is unfeasible for most consumer and emb…

View free PDFSource page
arxivcs.CRcs.AI2026-07-08

Closed-Loop Dynamic Validator Node Scaling in Private Substrate Blockchains Using Takagi-Sugeno Fuzzy Inference

Thandile Nododile, Ayinde M. Usman, Clement N. Nyirenda

Private blockchain networks run with fixed node configurations that cannot adapt to changing workload conditions. Too many nodes serving a light workload waste resources; too few nodes facing heavy demand slow block production and degrade finalisation. The right validator count i…

View free PDFSource page
arxivcs.CRcs.AI2026-06-25

ShareLock: A Stealthy Multi-Tool Threshold Poisoning Attack Against MCP

Liwei Liu, Tianzhu Han, Zijian Liu, Zishu Dong, Na Ruan

With the rapid evolution of LLM-driven agents, Model Context Protocol (MCP), an open protocol bridging LLMs with external tools, has quickly become foundational to modern agent ecosystems. However, the expanding adoption of MCP has also introduced novel security concerns such as…

View free PDFSource page