arXiv subjects

Evgenios M. Kornaropoulos

Publications and source records attributed to Evgenios M. Kornaropoulos.

15 recordsLinked to original sources

arXiv2026

TENNOR: Trustworthy Execution for Neural Networks through Obliviousness and Retrievals

Training neural networks on sensitive data in untrusted cloud environments requires simultaneously achieving computational efficiency and rigorous privacy guarantees. Sparsification techniques, essential for scalable training of wide layers, expose input-dependent memory-access patterns that are visible and can be exploited by a host OS/hypervisor, even when computation is protected by a Trusted Execution Environment. We present TENNOR, a system that resolves this tension by co-designing the neural network training pipeline with doubly oblivious primitives, eliminating value-dependent neural network access-pattern leakage while also utilizing adaptive sparsification. TENNOR recasts sparse neuron activation as a locality-sensitive hashing (LSH) retrieval problem, reducing secure sparsification to doubly oblivious accesses over an LSH data structure. To eliminate the prohibitive storage cost of "multi-table" LSH, we introduce Multi-Probe Winner-Take-All (MP-WTA): the first multi-probe scheme for rank-based LSH, achieving a $50\times$ reduction in (hash table) memory while preserving model accuracy. We evaluate TENNOR on multi-label classification benchmarks, with output layers of up to hundreds of thousands of neurons under different batch sizes inside an Intel TDX Trusted Domain, achieving per-batch training speedups of $13\times$-$470\times$ over a Path ORAM baseline and reducing a 208-hour run to about 26 minutes.

cs.CR↗

arXiv2026

HEAT: Faster Fully Homomorphic Inference via Approximations-Weights Co-Adaptation

Fully homomorphic encryption (FHE) allows a server to run a language model directly on encrypted user prompts, but current approaches remain prohibitively slow. Ciphertexts natively support only addition, multiplication, and rotation, and multiplications may be composed only to a bounded depth before a costly bootstrapping operation is required to continue. Every nonlinearity must therefore be approximated by an iterative method; each iteration increasing the number of multiplications. A higher iteration count buys precision but exhausts the available depth more frequently and thus triggers more bootstraps, which dominate latency. We introduce Homomorphic Encryption-Aware Training (HEAT), a fine-tuning method that makes the per-nonlinearity iteration counts learnable, enabling them and the model weights to co-adapt during training. HEAT optimizes iterations with respect to the task objective, allowing the model to adapt to approximation errors encountered during inference without architectural changes or retraining from scratch. We further relate iteration count to quantization bit width and bound, at fixed weights, the gap between our objective and quantization-aware training. On encrypted GPT-2 decoding, HEAT reduces iterations by $3.1\times$, bootstraps by $1.6\times$, and end-to-end latency by $1.4\times$, while improving decode agreement over the calibrated encrypted baseline.

cs.CR↗

arXiv2026

PILLAR: Private Inverted-Index Lexical Lookup for Augmented Retrieval

Retrieval-augmented generation (RAG) hands the user's query to whoever hosts the corpus. We propose PILLAR, a Privacy-Preserving RAG (PPRAG) system based on Private Information Retrieval (PIR) in which a client utilizes the k documents most similar to their query from a server-held and publicly known corpus to respond to their query, while the server learns nothing about the query, either its terms or its access pattern. Prior PPRAG constructions rely on dense retrieval alone, translating approximate nearest-neighbor search into many query-dependent rounds of PIR, and pay for it in both latency and retrieval quality. PILLAR instead performs private hybrid retrieval in two stages. A sparse stage issues a small, fixed number of PIR queries against a carefully designed index of precomputed BM25 scores, filtering the corpus down to candidates that share terms with the query without the server ever seeing which terms these are. A dense stage then fetches only those candidates' document embeddings and re-ranks them locally, avoiding the many costly PIR queries that private dense retrieval typically requires. We instantiate PILLAR with two protocols that trade latency against retrieval quality, each built on a different private rendering of lexical search. PILLAR-Bin bins posting lists into a hash table and is a single-round design that achieves lower latency than state-of-the-art private retrieval schemes. PILLAR-Tree turns block-max pruning into an oblivious tree traversal combined with cuckoo hash tables and achieves the highest retrieval quality at lower latency than state-of-the-art schemes.

cs.AI↗

arXiv2026

Safe Language Generation in the Limit

Recent results in learning a language in the limit have shown that, although language identification is impossible, language generation is tractable. As this foundational area expands, we need to consider the implications of language generation in real-world settings. This work offers the first theoretical treatment of safe language generation. Building on the computational paradigm of learning in the limit, we formalize the tasks of safe language identification and generation. We prove that under this model, safe language identification is impossible, and that safe language generation is at least as hard as (vanilla) language identification, which is also impossible. Last, we discuss several intractable and tractable cases.

cs.CL↗

arXiv2026

How Query Distribution Knowledge Breaks Multidimensional Encrypted Range Queries, With Guarantees

In this work, we show how knowledge of the query distribution, combined with access-pattern leakage, is sufficient to break multi-dimensional encrypted range queries, with provable guarantees. Prior attacks either recover only data topology without concrete coordinates for plaintexts (and as a result require post-hoc transformations), or assume adversarial control over database content; a strong and unrealistic threat model. Given knowledge of the query distribution, we revisit frequency matching, one of the earliest cryptanalytic ideas in this area, and push it to its limits in the multi-dimensional regime through LAMa ($\underline{L}$eakage-$\underline{A}$buse via $\underline{Ma}$tching). LAMa is a three-component framework that reconstructs plaintext coordinates in arbitrary dimensions without post-hoc transformations or data injection/poisoning. We complement LAMa with the first rigorous guarantees for multi-dimensional frequency-matching cryptanalysis, covering its query complexity, optimal parameterization, and worst-case reconstruction quality. Experiments on real-world data show that LAMa consistently outperforms the state of the art.

cs.CR↗

arXiv2025

Exposing Privacy Risks in Anonymizing Clinical Data: Combinatorial Refinement Attacks on k-Anonymity Without Auxiliary Information

Despite longstanding criticism from the privacy community, k-anonymity remains a widely used standard for data anonymization, mainly due to its simplicity, regulatory alignment, and preservation of data utility. However, non-experts often defend k-anonymity on the grounds that, in the absence of auxiliary information, no known attacks can compromise its protections. In this work, we refute this claim by introducing Combinatorial Refinement Attacks (CRA), a new class of privacy attacks targeting k-anonymized datasets produced using local recoding. This is the first method that does not rely on external auxiliary information or assumptions about the underlying data distribution. CRA leverages the utility-optimizing behavior of local recoding anonymization of ARX, which is a widely used open-source software for anonymizing data in clinical settings, to formulate a linear program that significantly reduces the space of plausible sensitive values. To validate our findings, we partnered with a network of free community health clinics, an environment where (1) auxiliary information is indeed hard to find due to the population they serve and (2) open-source k-anonymity solutions are attractive due to regulatory obligations and limited resources. Our results on real-world clinical microdata reveal that even in the absence of external information, established anonymization frameworks do not deliver the promised level of privacy, raising critical privacy concerns.

cs.CR↗

arXiv2025

When AIOps Become "AI Oops": Subverting LLM-driven IT Operations via Telemetry Manipulation

AI for IT Operations (AIOps) is transforming how organizations manage complex software systems by automating anomaly detection, incident diagnosis, and remediation. Modern AIOps solutions increasingly rely on autonomous LLM-based agents to interpret telemetry data and take corrective actions with minimal human intervention, promising faster response times and operational cost savings. In this work, we perform the first security analysis of AIOps solutions, showing that, once again, AI-driven automation comes with a profound security cost. We demonstrate that adversaries can manipulate system telemetry to mislead AIOps agents into taking actions that compromise the integrity of the infrastructure they manage. We introduce techniques to reliably inject telemetry data using error-inducing requests that influence agent behavior through a form of adversarial reward-hacking; plausible but incorrect system error interpretations that steer the agent's decision-making. Our attack methodology, AIOpsDoom, is fully automated--combining reconnaissance, fuzzing, and LLM-driven adversarial input generation--and operates without any prior knowledge of the target system. To counter this threat, we propose AIOpsShield, a defense mechanism that sanitizes telemetry data by exploiting its structured nature and the minimal role of user-generated content. Our experiments show that AIOpsShield reliably blocks telemetry-based attacks without affecting normal agent performance. Ultimately, this work exposes AIOps as an emerging attack vector for system compromise and underscores the urgent need for security-aware AIOps design.

cs.CR↗

arXiv2025

LLMmap: Fingerprinting For Large Language Models

We introduce LLMmap, a first-generation fingerprinting technique targeted at LLM-integrated applications. LLMmap employs an active fingerprinting approach, sending carefully crafted queries to the application and analyzing the responses to identify the specific LLM version in use. Our query selection is informed by domain expertise on how LLMs generate uniquely identifiable responses to thematically varied prompts. With as few as 8 interactions, LLMmap can accurately identify 42 different LLM versions with over 95% accuracy. More importantly, LLMmap is designed to be robust across different application layers, allowing it to identify LLM versions--whether open-source or proprietary--from various vendors, operating under various unknown system prompts, stochastic sampling hyperparameters, and even complex generation frameworks such as RAG or Chain-of-Thought. We discuss potential mitigations and demonstrate that, against resourceful adversaries, effective countermeasures may be challenging or even unrealizable.

cs.CR↗

arXiv2024

Hacking Back the AI-Hacker: Prompt Injection as a Defense Against LLM-driven Cyberattacks

Large language models (LLMs) are increasingly being harnessed to automate cyberattacks, making sophisticated exploits more accessible and scalable. In response, we propose a new defense strategy tailored to counter LLM-driven cyberattacks. We introduce Mantis, a defensive framework that exploits LLMs' susceptibility to adversarial inputs to undermine malicious operations. Upon detecting an automated cyberattack, Mantis plants carefully crafted inputs into system responses, leading the attacker's LLM to disrupt their own operations (passive defense) or even compromise the attacker's machine (active defense). By deploying purposefully vulnerable decoy services to attract the attacker and using dynamic prompt injections for the attacker's LLM, Mantis can autonomously hack back the attacker. In our experiments, Mantis consistently achieved over 95% effectiveness against automated LLM-driven attacks. To foster further research and collaboration, Mantis is available as an open-source tool: https://github.com/pasquini-dario/project_mantis

cs.CR↗

arXiv2024

Algorithmic Complexity Attacks on Dynamic Learned Indexes

Learned Index Structures (LIS) view a sorted index as a model that learns the data distribution, takes a data element key as input, and outputs the predicted position of the key. The original LIS can only handle lookup operations with no support for updates, rendering it impractical to use for typical workloads. To address this limitation, recent studies have focused on designing efficient dynamic learned indexes. ALEX, as the pioneering dynamic learned index structures, enables dynamism by incorporating a series of design choices, including adaptive key space partitioning, dynamic model retraining, and sophisticated engineering and policies that prioritize read/write performance. While these design choices offer improved average-case performance, the emphasis on flexibility and performance increases the attack surface by allowing adversarial behaviors that maximize ALEX's memory space and time complexity in worst-case scenarios. In this work, we present the first systematic investigation of algorithmic complexity attacks (ACAs) targeting the worst-case scenarios of ALEX. We introduce new ACAs that fall into two categories, space ACAs and time ACAs, which target the memory space and time complexity, respectively. First, our space ACA on data nodes exploits ALEX's gapped array layout and uses Multiple-Choice Knapsack (MCK) to generate an optimal adversarial insertion plan for maximizing the memory consumption at the data node level. Second, our space ACA on internal nodes exploits ALEX's catastrophic cost mitigation mechanism, causing an out-of-memory error with only a few hundred adversarial insertions. Third, our time ACA generates pathological insertions to increase the disparity between the actual key distribution and the linear models of data nodes, deteriorating the runtime performance by up to 1,641X compared to ALEX operating under legitimate workloads.

cs.DB↗

arXiv2022

The Price of Tailoring the Index to Your Data: Poisoning Attacks on Learned Index Structures

The concept of learned index structures relies on the idea that the input-output functionality of a database index can be viewed as a prediction task and, thus, be implemented using a machine learning model instead of traditional algorithmic techniques. This novel angle for a decades-old problem has inspired numerous exciting results in the intersection of machine learning and data structures. However, the main advantage of learned index structures, i.e., the ability to adjust to the data at hand via the underlying ML-model, can become a disadvantage from a security perspective as it could be exploited. In this work, we present the first study of poisoning attacks on learned index structures. The required poisoning approach is different from all previous works since the model under attack is trained on a cumulative distribution function (CDF) and, thus, every injection on the training set has a cascading impact on multiple data values. We formulate the first poisoning attacks on linear regression models trained on the CDF, which is a basic building block of the proposed learned index structures. We generalize our poisoning techniques to attack a more advanced two-stage design of learned index structures called recursive model index (RMI), which has been shown to outperform traditional B-Trees. We evaluate our attacks on real-world and synthetic datasets under a wide variety of parameterizations of the model and show that the error of the RMI increases up to $300\times$ and the error of its second-stage models increases up to $3000\times$.

cs.CR↗

arXiv2021

Adversarial Examples for $k$-Nearest Neighbor Classifiers Based on Higher-Order Voronoi Diagrams

Adversarial examples are a widely studied phenomenon in machine learning models. While most of the attention has been focused on neural networks, other practical models also suffer from this issue. In this work, we propose an algorithm for evaluating the adversarial robustness of $k$-nearest neighbor classification, i.e., finding a minimum-norm adversarial example. Diverging from previous proposals, we take a geometric approach by performing a search that expands outwards from a given input point. On a high level, the search radius expands to the nearby Voronoi cells until we find a cell that classifies differently from the input point. To scale the algorithm to a large $k$, we introduce approximation steps that find perturbations with smaller norm, compared to the baselines, in a variety of datasets. Furthermore, we analyze the structural properties of a dataset where our approach outperforms the competition.

cs.LG↗

arXiv2020

BeeTrace: A Unified Platform for Secure Contact Tracing that Breaks Data Silos

Contact tracing is an important method to control the spread of an infectious disease such as COVID-19. However, existing contact tracing methods alone cannot provide sufficient coverage and do not successfully address privacy concerns of the participating entities. Current solutions do not utilize the huge volume of data stored in business databases and individual digital devices. This information is typically stored in data silos and cannot be used due to regulations in place. To successfully unlock the potential of contact tracing, we need to consider both data utilization from multiple sources and the privacy of the participating parties. To this end, we propose BeeTrace, a unified platform that breaks data silos and deploys state-of-the-art cryptographic protocols to guarantee privacy goals.

cs.CR↗

arXiv2015

Optimizing Static and Adaptive Probing Schedules for Rapid Event Detection

We formulate and study a fundamental search and detection problem, Schedule Optimization, motivated by a variety of real-world applications, ranging from monitoring content changes on the web, social networks, and user activities to detecting failure on large systems with many individual machines. We consider a large system consists of many nodes, where each node has its own rate of generating new events, or items. A monitoring application can probe a small number of nodes at each step, and our goal is to compute a probing schedule that minimizes the expected number of undiscovered items at the system, or equivalently, minimizes the expected time to discover a new item in the system. We study the Schedule Optimization problem both for deterministic and randomized memoryless algorithms. We provide lower bounds on the cost of an optimal schedule and construct close to optimal schedules with rigorous mathematical guarantees. Finally, we present an adaptive algorithm that starts with no prior information on the system and converges to the optimal memoryless algorithms by adapting to observed data.

cs.DS↗

arXiv2011

Weak Dominance Drawings and Linear Extension Diameter

We introduce the problem of Weak Dominance Drawing for general directed acyclic graphs and we show the connection with the linear extension diameter of a partial order P. We present complexity results and bounds.

cs.DS↗