Decentralized finance (DeFi) platforms have experienced a rapid increase in fraudulent activities such as price manipulation, wash trading, and anomalous transaction behavior, while traditional centralized fraud detection methods remain unsuitable due to privacy and regulatory constraints. This study proposes a privacy-preserving fraud detection framework using federated learning, enabling multiple decentralized entities to collaboratively train a machine learning model without sharing raw transaction data. A real-world decentralized exchange (DEX) dataset containing over 100,000 transactions is preprocessed and enhanced through feature engineering techniques capturing swap rate deviations, transaction volume anomalies, and temporal patterns. In the absence of labeled fraud data, a heuristic-based labeling approach is employed to simulate realistic fraud scenarios. A Logistic Regression model is trained across multiple distributed client nodes, with model parameters aggregated using the Federated Averaging (FedAvg) algorithm over several communication rounds. The experimental findings show that the federated model delivers results similar to centralized methods while preserving data privacy, proving it to be an efficient solution for secure and scalable fraud detection in decentralized financial environments.
The secure, privacy-preserving sharing of genomic data across multiple institutions is a critical enabler for precision medicine, yet it remains fundamentally constrained by the identifiability and immutability of genomic data. While blockchain technologies have been proposed to provide decentralized governance, auditability, and tamper resistance for genomic data sharing, blockchain-only solutions are insufficient because they expose transaction metadata, access patterns, and smart-contract logic, leaving significant privacy risks unresolved. Zero-Knowledge Proofs (ZKPs) have recently emerged as a key cryptographic primitive for addressing such limitations, enabling verifiable access control, policy compliance, and computation correctness without disclosing sensitive genomic data. Although several surveys examine ZKPs or blockchain in isolation or across heterogeneous application domains, there is currently no dedicated survey that systematically analyzes their combined use in multi-party blockchain-based genomic data sharing systems. This paper addresses this gap by presenting a comprehensive, domain-specific survey of ZKP-enabled blockchain architectures for genomic data sharing. We classify existing approaches by architectural models, ZKP techniques, governance mechanisms, and threat-mitigation capabilities, and then compare their assumptions, performance characteristics, and deployment maturity. Furthermore, we identify open challenges in scalability, interoperability, proof overhead, and regulatory compliance, and outline future research directions for secure, scalable, and ethically compliant genomic data-sharing ecosystems.
We present CHRONOS, the first autonomous AI agent that simultaneously achieves plaintextblindness (all data is processed under fully homomorphic encryption without ever beingexposed), cryptographically enforced time bound existence (the agent’s own decryption key islocked behind a publicly verifiable proof of sequential work, rendering it inaccessible until aprecise future moment), and remote verifiability of self destruction (a zero knowledge proofcertifies that the key material has been irreversibly destroyed after mission completion). Theagent’s operational lifespan is governed by a “cryptographic fuse” constructed from a proof ofsequential work (PoSW) whose computation time accurately matches the intended missionduration. A drand decentralized randomness beacon serves as a trusted time oracle to trigger thefinal key shredding. Crucially, the erasure proof is a non interactive zero knowledge argument(SNARK) that proves the correct execution of the entire self destruction sequence—including thePoSW solution, decryption of the private key, and subsequent memory zeroization—enablingany third party to cryptographically verify the agent’s annihilation without trusting the agent orits hardware. We provide a complete system architecture, a formal security model with gamebased definitions and reductions to standard assumptions, and a proof of concept implementationusing Zama’s TFHE rs for encrypted inference, a Cohen Pietrzak PoSW implementation, and aGroth16 SNARK. Our benchmarks indicate that FHE inference on a small neural network (50 Kparameters) completes in seconds, the PoSW background thread consumes negligible resources,and the erasure proof can be generated and verified in under three seconds. CHRONOSrepresents a fundamental advance in secure, disposable AI agents, with immediate applications indefense, intelligence, and high privacy environments.
Zero-knowledge proof is a special cryptographic technique that allows a prover to convince a verifier about the correctness of a claim without explicitly disclosing the claim itself. With the advancements of blockchain technologies, zero-knowledge proof has been successfully integrated into many decentralized applications over the years. ZoKrates, with its ease-of-use and direct integration to blockchain platforms, has emerged as a leading framework for developing, generating and verifying zero-knowledge proofs. This survey compiles a corpus of 347 documents that cite the original research work of ZoKrates by considering the period ranging from 2018 to 2025. Out of this corpus, this survey systematically selects and analyzes a total of 87 different documents including only peer-reviewed publications and excluding the gray literature. To the best of our knowledge, this is the first survey in the literature to follow a systematic approach to analyze the privacy- preserving applications in blockchain from the perspective of ZoKrates. This survey presents three different classifications over the documents with respect to (i) the applications they develop, (ii) the challenges they frequently encounter and (iii) the metrics they often use to measure performance of their techniques. Based on the challenges identified, this survey finally discusses numerous future research directions to promote potential advancements in the field and attract the attention of scientific and industrial communities. Feedback from readers regarding any inaccuracies or misinformation in this survey is welcome.
With the rapid advancement of Web 3.0 technologies, public blockchain platforms are witnessing the emergence of novel services designed to enhance user privacy and anonymity. However, the powerful untraceability features inherent in these services inadvertently make them attractive tools for criminals seeking to launder illicit funds. Notably, existing de-anonymization methods face three major challenges when dealing with such transactions: highly homogenized transactional semantics, limited ability to model temporal discontinuities, and insufficient consideration of structural sparsity in account association graphs. To address these, we propose GradWATCH, designed to track anonymous accounts in Ethereum privacy-preserving services. Specifically, we first design a learnable account feature mapping module to extract informative transactional semantics from raw on-chain data. We then incorporate transaction relations into the account association graph to alleviate the adverse effects of structural sparsity. To capture temporal evolution, we further propose an edge-aware sliding-window mechanism that propagates and updates gradients at three granularities. Finally, we identify accounts controlled by the same entity by measuring their embedding distances in the learned representation space. Experimental results show that even under the conditions of unbalanced labels and sparse transactions, GradWATCH still achieves significant performance gains, with relative improvements ranging from 1.62% to 15. 22% in the MRR and from 3. 85% to 7. 31% in the F_1.
Digital identity verification often forces a privacy trade-off, where users must disclose sensitive personal data to prove simple eligibility criteria. As blockchain applications integrate with regulated environments, this over-disclosure creates significant risks of data breaches and surveillance. This work proposes a general Selective Disclosure Framework built on Ethereum, designed to decouple attribute verification from identity revelation. By utilizing client-side zk-SNARKs, the framework enables users to prove specific eligibility predicates without revealing underlying identity documents. We present a case study, ZK-Compliance, which implements a functional Grant, Verify, Revoke lifecycle for age verification. Preliminary results indicate that strict compliance requirements can be satisfied with negligible client-side latency (< 200 ms) while preserving the pseudonymous nature of public blockchains.
Bitcoin's block reward is scheduled to decline to zero, raising concerns about whether the network can remain secure once miners rely solely on transaction fees. This paper seeks to identify the conditions under which large-scale and persistent deviation from honest mining can arise. We analyze and compare the payoffs of honest and deviating miners in a sequential decision model, and identify a deviation threshold $G_t$ at which honest mining ceases to be privately optimal. Around the 2024 Bitcoin halving, we show that current mining behavior does not exhibit large-scale or structural deviation. However, when the block reward is removed, the $G_t$ criterion implies that deviation can arise even with a very small fraction of transaction fees. Finally, we evaluate three protocol-level mechanisms: Base Fee, Fee Floor, and an adaptive maximum block size rule, and show that their combination raises the deviation threshold and mitigates incentive breakdown in a fee-only regime. These results provide a practical benchmark for assessing Bitcoin's security as block rewards disappear.
Automated negotiations in insurance and business-to-business (B2B) commerce encounter substantial challenges. Current systems force a trade-off between convenience and privacy by routing sensitive financial data through centralized servers, increasing security risks, and diminishing user trust. This study introduces a device-native autonomous Artificial Intelligence (AI) agent system for privacy-preserving negotiations. The proposed system operates exclusively on user hardware, enabling real-time bargaining while maintaining sensitive constraints locally. It integrates zero-knowledge proofs to ensure privacy and employs distilled world models to support advanced on-device reasoning. The architecture incorporates six technical components within an agentic AI workflow. Agents autonomously plan negotiation strategies, conduct secure multi-party bargaining, and generate cryptographic audit trails without exposing user data to external servers. The system is evaluated in insurance and B2B procurement scenarios across diverse device configurations. Results show an average success rate of 87%, a 2.4x latency improvement over cloud baselines, and strong privacy preservation through zero-knowledge proofs. User studies show 27% higher trust scores when decision trails are available. These findings establish a foundation for trustworthy autonomous agents in privacy-sensitive financial domains.
Pankaj Kumar, Arun K H, Yogesh N, Prakash Babu · 8 authors
The growing dependance on data-based decisionmaking in healthcare has brought attention to the vital importance of secure, privacy-preserving and collaborative learning techniques. Traditional centralized learning approaches in medical data often raise concerns regarding patient privacy data leaks and even regulatory troubles. Federated learning came as a good solution, where it allows model training in different hospitals without sharing the sensitive patient data. However, federated learning has its problems - it can be mislead with fake updates, the model can even be poisoned and it is really hard to trust every participants involved. In this work, we present fed-chain, a secure and scalable framework which brings together federated learning, blockchain and zero-knowledge-proofs(ZKPs) preserving the privacy of patient's data in healthcare. Blockchain here adds decentralized trust, immutability and makes model updates transparent to review while ZKPs helps in proving correctness without leaking personal data. We are implemented this framework for heart disease prediction where multiple hospitals train the model together but the data stays confidential. Our experimental results shown better accuracy, more strength against attacks and even low communication cost compared to other FL setups. Overall, the systems gives a safer approach for working together on healthcare data, allowing hospitals and research centers to generate valuable predictions using these models while keeping the patient data private and safe.
The fast growth of the Internet of Medical Things (IoMT) has made it much harder to manage and protect medical data because of security and privacy issues. This paper suggests a secure federated cloud storage system that uses a hybrid heuristic attribute-based encryption (ABE) scheme combined with a permissioned Blockchain to solve these problems. The suggested system improves data privacy and integrity by first gathering medical data and then encrypting it with ABE using the best key made by the Hybrid Mexican Axolotl with Energy Valley Optimizer (HMO-EVO). A permissioned blockchain securely stores the encrypted data, making sure that access is tightly controlled and that data breaches are avoided. The system uses federated learning with a Multi-scale Bi-Long Short-Term Memory and Gated Recurrent Unit (MBiLSTM-GRU) to make accurate predictions about diseases. This helps with healthcare monitoring. This federated approach lets deep learning models be trained in different places, keeping patient data private while still allowing for collective learning. The experimental results demonstrate that the proposed system surpasses traditional methods regarding security, efficiency, and predictive accuracy. This study presents an extensive framework for the secure management of medical data, integrating the advantages of federated learning and blockchain technology to tackle the essential challenges of data ownership, regulatory adherence, and privacy within IoMT networks.
In Vertical Federated Learning (VFL), prior work has primarily focused on protecting data privacy, while overlooking the risk that participants may manipulate local model execution to mount integrity attacks.Integrating zero-knowledge proofs (ZKPs) into the training process can ensure that each party's computations are verifiable without revealing private data.However, directly encoding deep model training as a monolithic ZKP circuit is impractical due to: (i) complex circuit design and high overhead from frequent parameter commitments, (ii) expensive proof generation for embeddings(cross-party information interface), and (iii) synchronous proof generation that blocks iterative training rounds.To address these challenges, we present ZKSL, an efficient and asynchronous VFL framework that achieves verifiable training under a malicious threat model.ZKSL partitions deep neural networks into layer-wise circuits and generates their proofs in parallel, ensuring input-output consistency via Privacy-Commitment PLONK (PC-PLONK), a lightweight extension that supports low-cost, iteration-by-iteration parameter commitments.For embedding layers, ZKSL adopts a probabilistic verification technique that reduces proof complexity from O(N nd) to O(nd).Furthermore, ZKSL incorporates an asynchronous compute-prove scheduling mechanism to decouple proof generation from training iterations, effectively mitigating pipeline stalls.Experimental results on DeepFM and CNN models show that ZKSL reduces proof generation time by up to 73% while maintaining 99.4% accuracy, demonstrating superior scalability and practicality for real-world federated learning.
In the digital economy era, the contradiction between data sharing and privacy protection is increasingly prominent. Traditional privacy protection technologies such as anonymization, differential privacy, and homomorphic encryption are difficult to meet the diverse privacy demands in multiple fields due to their flaws such as vulnerability to cracking, sacrificing data accuracy, and high computational complexity. Zero-knowledge proof (ZKP), with its core advantage of "data available but not visible", has become the key technical support to solve this contradiction. However, the current protocol types are complex and the demands in various fields vary significantly, leading to confusion in protocol selection and imbalance in resource allocation in practice. This paper systematically reviews the theoretical basis, technical system and mainstream protocol characteristics of zero-knowledge proof, integrates the advantages of demand-driven decision tree and protocol three-dimensional classification model, constructs a protocol selection framework of "demand-driven - feature matching - dynamic optimization", and clarifies the multi-stage decision-making process; Four typical fields, namely education, machine learning, finance, and healthcare, were selected to verify the effectiveness and universality of the framework, providing standardized tools for the large-scale application of zero-knowledge proof technology in various fields and direction guidance for the development of privacy protection technology in the post-quantum era.
Federated learning (FL) enables collaborative model training while preserving data privacy, yet both centralized and decentralized approaches face challenges in scalability, security, and update validation. We propose ZK-HybridFL, a secure decentralized FL framework that integrates a directed acyclic graph (DAG) ledger with dedicated sidechains and zero-knowledge proofs (ZKPs) for privacy-preserving model validation. The framework uses event-driven smart contracts (EDSCs) and an oracle-assisted sidechain to verify local model updates without exposing sensitive data. A built-in challenge mechanism efficiently detects adversarial behavior. In experiments on image classification and language modeling tasks, ZK-HybridFL achieves faster convergence, higher accuracy, lower perplexity, and reduced latency compared to Blade-FL and ChainFL. It remains robust against substantial fractions of adversarial and idle nodes, supports sub-second on-chain verification with efficient gas usage, and prevents invalid updates and orphanage-style attacks. This makes ZK-HybridFL a scalable and secure solution for decentralized FL across diverse environments.
Large-scale privacy-preserving computation is essential for cross-domain data collaboration, making correctness auditing and accountability crucial for practical deployment. Existing auditing schemes typically use zero-knowledge proofs (ZKPs) to verify computation correctness and blockchain-based stake or reputation mechanisms to constrain dishonest participants. However, ZKP approaches often verify computation tasks or circuit constraints independently, causing high communication and verification overhead in large-scale task scenarios. Meanwhile, existing blockchain-based accountability mechanisms are weakly coupled with cryptographic audit results and lack sufficient adaptability in dynamic environments. In this paper, we propose MTBA, a batch auditing framework for privacy-preserving computation that leverages VOLE correlation to support polynomial proof generation and integrates blockchain-based reputation feedback. MTBA transforms arithmetic-circuit computations into auditable polynomial tasks by encoding multiplication-gate consistency as polynomial relations, converts these tasks into compact polynomial proofs, and introduces polynomial proof aggregation at multi-task for batch auditing. It further records signed audit outcomes on blockchain to support adaptive reputation feedback and malicious party accountability. Experimental results on circuits with up to $10^{7}$ multiplication gates show that MTBA improves auditing throughput by up to 2.7 times faster than conventional independent auditing schemes. These results indicate that MTBA provides a scalable and accountable auditing mechanism for large-scale privacy-preserving computation.
In the face of the regulatory failure problem caused by blockchain hidden addresses, existing solutions often fall into a dilemma where 'privacy protection' and 'compliance review' are either one or the other.This paper proposes an innovative integration framework that transforms the behavioural elements in anti-money laundering and other legal provisions (such as 'high-frequency and small-scale transactions') into computable logic.Based on zero-knowledge proof technology, it generates verifiable credentials to determine whether the transaction behaviour is compliant without revealing the true identity of the address.Experiments on a public blockchain transaction dataset (elliptic) show that this framework achieves an average improvement of over 15% in core identification performance compared to traditional non-private rule-based methods, while maintaining an acceptable performance overhead.As a proof-of-concept validation conducted on a transparent dataset with simulated concealment, the actual performance may differ in native privacy-preserving chains.This research provides a new approach that combines legal rigor with technical feasibility for achieving effective on-chain behaviour supervision while protecting user privacy.
Federatívne učenie (FL) umožňuje spoločné trénovanie modelu bez priameho zdieľania údajov, ale často sa spolieha na silné predpoklady o čestnom správaní klienta a servera. To je dôvod, prečo štandardné FL protokoly poskytujú iba obmedzenú záruku ohľadom výpočtov na strane klienta, integrity odoslaných informácii, alebo ohľadom správnosti agregácie na strane servera. Táto diplomová práca skúma použitie systémov s nulovými znalosťami (ZKP) spolu s podpornými metódami na vytvorenie dôvery v FL prostredí. V tejto práci sa po úvode k FL a ZKP ďalej skúma prehľad existujúcich ZKP nástrojov v prostredí FL. Na základe tejto analýzy je vytvorená kategorizácia existujúcich prístupov FL založených na ZKP, ktorá je postavená najmä na cieľoch daného systému. Na základe identifikovaných možností zlepšenia práca navrhuje overiteľný protokol váženej agregácie. V tomto protokole je každý prijatý príspevok previazaný s autorizovanou váhou, prípustnou skrytou aktualizáciou, konzistentným váženým vstupom a výslednou aktualizáciou modelu, ktorú je možné verejne overiť prepočítaním. Tento protokol bol implementovaný ako prototyp s plne funkčnými kryptografickými komponentami. Následne je tento protokol vyhodnotený.
Agent-based federated learning (FL) enables multiple nodes to collaboratively train a global model without sharing raw data, thereby mitigating privacy and security concerns. However, FL inherently relies on the credibility of participating agents, making trust assurance critical in environments susceptible to malicious attacks and operational faults. This paper proposes a procedural architecture for evaluating and enhancing agent credibility in distributed ledger-based FL (DLFL) systems. The proposed architecture spans the agent life cycle, consisting of the pre-training, in-training, and post-training stages, and integrates three verification domains: data-based, behavior-based, and technology-based verification. In the pre-training stage, data-based verification assesses data integrity and quality through reference distribution analysis and entropy reduction to improve learning stability. During the in-training stage, behavior-based verification establishes a dual mechanism that monitors and analyzes agents’ learning processes and outcomes and detects anomalies in updates and performance. In the post-training stage, technology-based verification ensures record immutability and accountability through blockchain, cryptographic validation, and auditing mechanisms. By combining these multi-layered procedures, the proposed architecture enables systematic and continuous evaluation of agent credibility, fostering a trustworthy FL ecosystem and enabling future applications in autonomous agent collaboration and trust-oriented AI governance.
Deny Prasetyo, Siska Narulita, Ahmad Jurnaidi Wahidin, Rosalina Yani Widiastuti · 7 authors
This study introduces a trust centric machine learning framework designed to improve decision making reliability and security in decentralized digital service ecosystems. Traditional machine learning models often focus on accuracy and efficiency but fail to address the challenges of trust and security in decentralized environments. In contrast, the proposed framework integrates dynamic trust indicators and employs Federated Learning (FL) to ensure privacy while enhancing decision making performance. The framework also incorporates Zero Knowledge Proofp based Verifiable Machine Learning (ZKP-VML), which ensures transparency and security without compromising sensitive data. Through continuous real time trust assessments, the framework adapts to changing conditions, improving the accuracy and reliability of decisions in environments where participants may not fully trust each other. The application of this framework in autonomous vehicles and IoT networks demonstrated its ability to make robust, secure decisions, even in complex and uncertain scenarios. The framework’s ability to incorporate both trust and security into its decision making processes sets it apart from traditional models, which typically do not address the trustworthiness of data or participants. This research highlights the importance of integrating trust and security into machine learning models, particularly in decentralized systems, and offers a robust solution to trust management challenges. However, challenges such as scalability and computational efficiency remain, and future work should focus on enhancing these aspects, along with exploring the framework's applicability in other decentralized domains like finance or supply chain management. The integration of privacy preserving technologies and improvements in adversarial robustness are also potential areas for future research.
Abdullah Ayub Khan, Asif Ali Laghari, Hamad Almansour, Teerath Kumar · 7 authors
Wearable health technology has revolutionized remote monitoring and personalized healthcare by allowing real-time surveillance of patient health measurements and vital signs. However, their widespread acceptance is hampered by issues with security, privacy preservation, data protection, and interoperability. Blockchain Technology (BT), in particular Zero-Knowledge Proofs (ZKPs) and smart contracts, present a viable way to enhance the privacy, provenance, and integrity of wearable health data. This paper proposes a BT-enabled system that guarantees decentralized, unforged data management, transparency, immutability, and dynamic traceability for wearable health devices, particularly smartwatches with biosensors. To evaluate the effectiveness of the proposed work, the main performance-related metrics-latency, throughput, computational overhead, security robustness, and scalability-are looked at. The experiment's simulated findings show that BT integration is effective, with a 99.33% improvement in data integrity and protection. Automated access control protocols demonstrate data protection by utilizing smart contracts, and ZKPs guarantee verifiable data exchanges without jeopardizing patient privacy. These results demonstrate improved interoperability, decreased processing time, and increased security in comparison to comparable cutting-edge centralized platforms.
We present the Mazari Bidirectional Architecture, a cryptographic framework providing end-to-end privacy guarantees for computation systems employing homomorphic encryption (HE), differential privacy (DP), and zero-knowledge proofs (ZK). The architecture comprises two complementary orderings: Y.I.N. Mazari Ordering (DP→ZK→HE) for secure data submission and Y.A.N.G. Mazari Ordering (VERIFY→DISAGGREGATE→DECRYPT→DENOISE) for secure result retrieval. Through exhaustive permutation analysis, we prove that among all possible orderings for each direction, exactly one ordering in each direction is secure—and these orderings are mathematical duals. The Y.I.N. ordering prevents information leakage during data submission, while the Y.A.N.G. ordering prevents timing side-channel attacks during result retrieval. This paper provides comprehensive coverage of all implementation variations, security proofs, and robustness analysis to establish technical foundations for privacy-preserving computation research.
Modern cloud-based AI training relies on extensive telemetry and logs to ensure accountability. While these audit trails enable retrospective inspection, they struggle to address the inherent non-determinism of deep learning. Stochastic operations, such as dropout, create an ambiguity surface where attackers can mask malicious manipulations as natural random variance, granting them plausible deniability. Consequently, existing logging mechanisms cannot verify whether stochastic values were generated and applied honestly without exposing sensitive training data. To close this integrity gap, we introduce Verifiable Dropout, a privacy-preserving mechanism based on zero-knowledge proofs. We treat stochasticity not as an excuse but as a verifiable claim. Our approach binds dropout masks to a deterministic, cryptographically verifiable seed and proves the correct execution of the dropout operation. This design enables users to audit the integrity of stochastic training steps post-hoc, ensuring that randomness was neither biased nor cherry-picked, while strictly preserving the confidentiality of the model and data.