arXiv subjects

Gregory Blanc

Publications and source records attributed to Gregory Blanc.

6 recordsLinked to original sources

arXiv2026

BackDFL: A Unified Benchmark For Backdoor Attacks and Defenses In Decentralized Federated Learning

Decentralized Federated Learning (DFL) promises trust-free collaborative learning by replacing the centralized parameter server with peer-to-peer model exchange. However, this architectural shift fundamentally reshapes the threat landscape. Without globally coordinated aggregation, DFL becomes particularly susceptible to backdoor attacks, in which malicious participants implant persistent hidden behaviors while maintaining high clean-task performance. In this paper, we argue that the robustness of DFL has been significantly overestimated. Existing studies rely on simplified threat models, non-adaptive adversaries, fragmented evaluation protocols, inconsistent communication topologies, and ad hoc training configurations, leading to an incomplete understanding of DFL security. To address these limitations, we present BackDFL, a unified benchmark for systematically evaluating DFL under realistic and adaptive backdoor attacks. Through extensive experiments, BackDFL exposes critical failure modes of decentralized learning. Our results demonstrate that both state-of-the-art Byzantine-robust DFL methods and adapted FL backdoor defenses fail under modest malicious participation rates (as low as 15%), especially in heterogeneous settings, while their robustness varies substantially across communication graph topologies.

cs.LG↗

arXiv2026

Robust Reputation-Driven Crowdsourced Federated Learning

Crowdsourced Federated Learning (CrowdFL) extends traditional federated learning by enabling open and heterogeneous participation through a crowdsourcing paradigm. In this setting, reputation-driven incentive mechanisms are commonly employed to guide worker selection and enhance trustworthiness. While such approaches improve participant reliability, existing frameworks largely overlook the quantification of their robustness against stealthy adversaries, particularly those capable of evading standard detection mechanisms. To fill this gap, this paper proposes R2CFL, a robust reputation-driven CrowdFL framework. R2CFL introduces a robust reputation model coupled with a nearest neighbor mixing (R2-NNM) defense mechanism that links reputation evolution with the filtering of updates during aggregation. The proposed mechanism prevents stealthy attackers from gradually accumulating trust and influencing future tasks. Experimental results demonstrate that R2-NNM matches or surpasses state-of-the-art Byzantine-robust and backdoor defense mechanisms against adaptive attackers. Furthermore, when integrated with existing detect-and-filter defenses, the proposed reputation model faithfully captures the statistical robustness of the underlying defense by producing reputation scores that closely reflect its true positive and false positive characteristics.

cs.LG↗

arXiv2026

TACTIC-KG: Toward Small Agent Teams for Cyber Threat Intelligence Knowledge Graph Construction

Cyber Threat Intelligence (CTI) reports are predominantly unstructured, heterogeneous, and noisy, which limits their direct usability for automated analysis and reasoning. Cybersecurity Knowledge Graphs (CSKGs) provide a structured representation of adversarial entities, actions, and relations, but constructing such graphs from free-text CTI remains a challenge. Recent approaches rely on monolithic Large Language Models (LLMs) to perform end-to-end extraction and completion, leading to high cost, limited controllability, and unstable performance. This paper introduces TACTIC-KG, an agentic framework for CSKG construction that decomposes the task into modular, specialized LLM agents responsible for extraction, typing, verification, and curation. Using lightweight models (3B--8B), TACTIC-KG improves stability, recall, and graph consistency while reducing deployment cost. We implement and evaluate TACTIC-KG against recent state-of-the-art systems. Experiments on human-annotated CTI reports show that agent specialization consistently outperforms larger monolithic in-context-learning (ICL) baselines in extraction F1-score, typing accuracy, and structural graph similarity.

cs.CR↗

arXiv2025

PROTEAN: Federated Intrusion Detection in Non-IID Environments through Prototype-Based Knowledge Sharing

In distributed networks, participants often face diverse and fast-evolving cyberattacks. This makes techniques based on Federated Learning (FL) a promising mitigation strategy. By only exchanging model updates, FL participants can collaboratively build detection models without revealing sensitive information, e.g., network structures or security postures. However, the effectiveness of FL solutions is often hindered by significant data heterogeneity, as attack patterns often differ drastically across organizations due to varying security policies. To address these challenges, we introduce PROTEAN, a Prototype Learning-based framework geared to facilitate collaborative and privacy-preserving intrusion detection. PROTEAN enables accurate detection in environments with highly non-IID attack distributions and promotes direct knowledge sharing by exchanging class prototypes of different attack types among participants. This allows organizations to better understand attack techniques not present in their data collections. We instantiate PROTEAN on two cyber intrusion datasets collected from IIoT and 5G-connected participants and evaluate its performance in terms of utility and privacy, demonstrating its effectiveness in addressing data heterogeneity while improving cyber attack understanding in federated intrusion detection systems (IDSs).

cs.CR↗

arXiv2019

Anomalous Communications Detection in IoT Networks Using Sparse Autoencoders

Nowadays, IoT devices have been widely deployed for enabling various smart services, such as, smart home or e-healthcare. However, security remains as one of the paramount concern as many IoT devices are vulnerable. Moreover, IoT malware are constantly evolving and getting more sophisticated. IoT devices are intended to perform very specific tasks, so their networking behavior is expected to be reasonably stable and predictable. Any significant behavioral deviation from the normal patterns would indicate anomalous events. In this paper, we present a method to detect anomalous network communications in IoT networks using a set of sparse autoencoders. The proposed approach allows us to differentiate malicious communications from legitimate ones. So that, if a device is compromised only malicious communications can be dropped while the service provided by the device is not totally interrupted. To characterize network behavior, bidirectional TCP flows are extracted and described using statistics on the size of the first N packets sent and received, along with statistics on the corresponding inter-arrival times between packets. A set of sparse autoencoders is then trained to learn the profile of the legitimate communications generated by an experimental smart home network. Depending on the value of N, the developed model achieves attack detection rates ranging from 86.9% to 91.2%, and false positive rates ranging from 0.1% to 0.5%.

cs.CR↗

arXiv2014

Combining Technical and Financial Impacts for Countermeasure Selection

Research in information security has generally focused on providing a comprehensive interpretation of threats, vulnerabilities, and attacks, in particular to evaluate their danger and prioritize responses accordingly. Most of the current approaches propose advanced techniques to detect intrusions and complex attacks but few of these approaches propose well defined methodologies to react against a given attack. In this paper, we propose a novel and systematic method to select security countermeasures from a pool of candidates, by ranking them based on the technical and financial impact associated to each alternative. The method includes industrial evaluation and simulations of the impact associated to a given security measure which allows to compute the return on response investment for different candidates. A simple case study is proposed at the end of the paper to show the applicability of the model.

cs.CR↗