Artificial intelligence (AI) is increasingly central to solving complex societal, economic, and scientific problems, yet prevailing models remain constrained by their opacity, vulnerability to adversarial inputs, and reliance on centralized infrastructures. These limitations underscore the urgent need for approaches that combine the adaptability of neural networks with the interpretability and rule-based precision of symbolic systems. At the same time, decentralization has emerged as a critical paradigm for enhancing trust, resilience, and accountability in intelligent systems. Together, these threads converge on the concept of decentralized neuro-symbolic cognitive systems, which integrate distributed inference, symbolic reasoning, and governance mechanisms to create secure and transparent frameworks for machine intelligence. This article presents a comprehensive methodology for the design and operation of such systems, advancing beyond conventional hybrid AI by embedding causal intent routing, federated cognitive capsules, encrypted episodic memory, and immutable epistemic ledgers. These elements are supported by governance innovations such as the NeuroConstitutionâ„¢, which enables tokenized, evolvable norms and ensures accountability through transparent dispute resolution. The framework is evaluated across key application domains, including healthcare, finance, governance, and climate modeling, with comparative benchmarks demonstrating gains in robustness, interpretability, and systemic trust. By uniting symbolic reasoning, neural inference, and decentralized governance, this research outlines a pathway toward AI systems that are not only technically powerful but also socially aligned and ethically defensible. The article concludes that decentralized neuro-symbolic architectures provide a sustainable foundation for advancing trustworthy AI capable of supporting critical infrastructures and decision-making in a rapidly evolving world.
Yating Liu, Xing Su, Hao Wu, Sijin Li · 7 authors
Adversarial smart contracts, mostly on EVM-compatible chains like Ethereum and BSC, are deployed as EVM bytecode to exploit vulnerable smart contracts for financial gain. Detecting such malicious contracts at the time of deployment is an important proactive strategy to prevent losses from victim contracts. It offers a better cost-benefit ratio than detecting vulnerabilities on diverse potential victims. However, existing works are not generic with limited detection types and effectiveness due to imbalanced samples, while the emerging LLM technologies, which show their potential in generalization, have two key problems impeding its application in this task: hard digestion of compiled-code inputs, especially those with task-specific logic, and hard assessment of LLM's certainty in its binary (yes-or-no) answers. Therefore, we propose a generic adversarial smart contracts detection framework FinDet, which leverages LLM with two enhancements addressing the above two problems. FinDet takes as input only the EVM bytecode contracts and identifies adversarial ones among them with high balanced accuracy. The first enhancement extracts concise semantic intentions and high-level behavioral logic from the low-level bytecode inputs, unleashing the LLM reasoning capability restricted by the task input. The second enhancement probes and measures the LLM uncertainty to its multi-round answering to the same query, improving the LLM answering robustness for binary classifications required by the task output. Our comprehensive evaluation shows that FinDet achieves a BAC of 0.9374 and a TPR of 0.9231, significantly outperforming existing baselines. It remains robust under challenging conditions including unseen attack patterns, low-data settings, and feature obfuscation. FinDet detects all 5 public and 20+ unreported adversarial contracts in a 10-day real-world test, confirmed manually.
Split Learning (SL) is a distributed learning approach that enables resource-constrained clients to collaboratively train deep neural networks (DNNs) by offloading most layers to a central server while keeping in- and output layers on the client-side. This setup enables SL to leverage server computation capacities without sharing data, making it highly effective in resource-constrained environments dealing with sensitive data. However, the distributed nature enables malicious clients to manipulate the training process. By sending poisoned intermediate gradients, they can inject backdoors into the shared DNN. Existing defenses are limited by often focusing on server-side protection and introducing additional overhead for the server. A significant challenge for client-side defenses is enforcing malicious clients to correctly execute the defense algorithm. We present ZORRO, a private, verifiable, and robust SL defense scheme. Through our novel design and application of interactive zero-knowledge proofs (ZKPs), clients prove their correct execution of a client-located defense algorithm, resulting in proofs of computational integrity attesting to the benign nature of locally trained DNN portions. Leveraging the frequency representation of model partitions enables ZORRO to conduct an in-depth inspection of the locally trained models in an untrusted environment, ensuring that each client forwards a benign checkpoint to its succeeding client. In our extensive evaluation, covering different model architectures as well as various attack strategies and data scenarios, we show ZORRO's effectiveness, as it reduces the attack success rate to less than 6\% while causing even for models storing \numprint{1000000} parameters on the client-side an overhead of less than 10 seconds.
Nan Wang, Nan Wu, Xiangyu Hui, Jiafan Wang · 5 authors
As the demand for exercising the "right to be forgotten" grows, the need for verifiable machine unlearning has become increasingly evident to ensure both transparency and accountability. We present {\em zkUnlearner}, the first zero-knowledge framework for verifiable machine unlearning, specifically designed to support {\em multi-granularity} and {\em forgery-resistance}. First, we propose a general computational model that employs a {\em bit-masking} technique to enable the {\em selectivity} of existing zero-knowledge proofs of training for gradient descent algorithms. This innovation enables not only traditional {\em sample-level} unlearning but also more advanced {\em feature-level} and {\em class-level} unlearning. Our model can be translated to arithmetic circuits, ensuring compatibility with a broad range of zero-knowledge proof systems. Furthermore, our approach overcomes key limitations of existing methods in both efficiency and privacy. Second, forging attacks present a serious threat to the reliability of unlearning. Specifically, in Stochastic Gradient Descent optimization, gradients from unlearned data, or from minibatches containing it, can be forged using alternative data samples or minibatches that exclude it. We propose the first effective strategies to resist state-of-the-art forging attacks. Finally, we benchmark a zkSNARK-based instantiation of our framework and perform comprehensive performance evaluations to validate its practicality.
Decentralized Finance (DeFi) attacks have resulted in significant losses, often orchestrated through Adversarial Exploiter Contracts (AECs) that exploit vulnerabilities in victim smart contracts. To proactively identify such threats, this paper targets the explainable detection of AECs. Existing detection methods struggle to capture semantic dependencies and lack interpretability, limiting their effectiveness and leaving critical knowledge gaps in AEC analysis. To address these challenges, we introduce SEASONED, an effective, self-explanatory, and robust framework for AEC detection. SEASONED extracts semantic information from contract bytecode to construct a semantic relation graph (SRG), and employs a self-counterfactual explainable detector (SCFED) to classify SRGs and generate explanations that highlight the core attack logic. SCFED further enhances robustness, generalizability, and data efficiency by extracting representative information from these explanations. Both theoretical analysis and experimental results demonstrate the effectiveness of SEASONED, which showcases outstanding detection performance, robustness, generalizability, and data efficiency learning ability. To support further research, we also release a new dataset of 359 AECs.
Fine-tuning large language models (LLMs) is crucial for adapting them to specific tasks, yet it remains computationally demanding and raises concerns about correctness and privacy, particularly in untrusted environments. Although parameter-efficient methods like Low-Rank Adaptation (LoRA) significantly reduce resource requirements, ensuring the security and verifiability of fine-tuning under zero-knowledge constraints remains an unresolved challenge. To address this, we introduce VeriLoRA, the first framework to integrate LoRA fine-tuning with zero-knowledge proofs (ZKPs), achieving provable security and correctness. VeriLoRA employs advanced cryptographic techniques -- such as lookup arguments, sumcheck protocols, and polynomial commitments -- to verify both arithmetic and non-arithmetic operations in Transformer-based architectures. The framework provides end-to-end verifiability for forward propagation, backward propagation, and parameter updates during LoRA fine-tuning, while safeguarding the privacy of model parameters and training data. Leveraging GPU-based implementations, VeriLoRA demonstrates practicality and efficiency through experimental validation on open-source LLMs like LLaMA, scaling up to 13 billion parameters. By combining parameter-efficient fine-tuning with ZKPs, VeriLoRA bridges a critical gap, enabling secure and trustworthy deployment of LLMs in sensitive or untrusted environments.
Youwei Huang, Jianwen Li, Bin Hu, Sen Fang · 6 authors
Malicious developer intents in smart contracts constitute significant security threats to decentralized applications, leading to substantial economic losses. Prior work introduced SmartIntentNN, a deep learning model for detecting unsafe developer intents. By combining the Universal Sentence Encoder, a K-means clustering-based intent highlighting mechanism, and a Bidirectional Long Short-Term Memory (BiLSTM) network, the model achieved an F1 score of 0.8633 on an evaluation set of 10,000 real-world smart contracts across ten distinct intent categories. This paper presents SmartIntentV2 (Smart Contract Intent Neural Network Version 2). The primary enhancement is the integration of a BERT-based pre-trained programming language model, which we domain-adaptively pre-train on a dataset of 16,000 real-world smart contracts using a Masked Language Modeling objective. SmartIntentV2 retains the BiLSTM-based multi-label classification network for intent detection. On the same evaluation set of 10,000 smart contracts, it achieves superior performance with an accuracy of 0.9789, precision of 0.9090, recall of 0.9476, and an F1 score of 0.9279, substantially outperforming its predecessor and other baseline models. Notably, SmartIntentV2 also delivers a 65.5% relative improvement in F1 score over GPT-4.1 on this specialized task. These results establish SmartIntentV2 as a new state-of-the-art model for smart contract intent detection.
Amjad Almaghthawi, Wael M.S. Yafooz, Nasser S. Albalawi
In decentralized apps, smart contracts are used to conduct trusted transactions on the Blockchain (BC). While smart contracts are highly effective, they are also highly susceptible to security flaws, leading to serious financial consequences. However, the combination of BC technology and artificial intelligence provides a solution for powerful, secure, and decentralized applications in various sectors. Furthermore, large language models (LLMs), which are essential advanced machine learning frameworks, are now used in various applications, including customer service, chatbots, code generation, vulnerability detection, and language translation. This study investigates the use of LLMs for automated vulnerability detection in Solidity-based smart contracts. Specifically, three models are evaluated and compared: GPT-3.5-turbo, DeepSeek R1, and LLaMA-3. With a labeled, multi-class dataset including four vulnerability types, the models are assessed across three reasoning strategies: zero-shot, few-shot, and chain of thought. A prompt-based evaluation and performance comparison is conducted using standard metrics such as accuracy, precision, recall, F1-score, and average detection time. Results show that in the zero-shot setting, GPT-3.5-turbo achieves the highest accuracy of 94.59%, followed closely by LLaMA-3 with 92%, while DeepSeek R1 achieved 78.95%. In the few-shot setting, LLaMA-3 outperformed other models. Furthermore, in the CoT setting, LLaMA-3 demonstrates the strongest overall performance with 96% accuracy and an F1-score of 0.82, surpassing DeepSeek R1's average of 78.95% and GPT-3.5's CoT performance, which is notably lower. Hence, this study develops an evaluation framework for LLM-based vulnerability detection, and we have demonstrated that prompt engineering has the potential to enhance the security of smart contracts.
Dan Ivanov, Tristan Freiberg, Shahabi, Shirin, Jonathan Gold · 5 authors
DSperse is a modular framework for distributed machine learning inference with strategic cryptographic verification. Operating within the emerging paradigm of distributed zero-knowledge machine learning, DSperse avoids the high cost and rigidity of full-model circuitization by enabling targeted verification of strategically chosen subcomputations. These verifiable segments, or "slices", may cover part or all of the inference pipeline, with global consistency enforced through audit, replication, or economic incentives. This architecture supports a pragmatic form of trust minimization, localizing zero-knowledge proofs to the components where they provide the greatest value. We evaluate DSperse using multiple proving systems and report empirical results on memory usage, runtime, and circuit behavior under sliced and unsliced configurations. By allowing proof boundaries to align flexibly with the model's logical structure, DSperse supports scalable, targeted verification strategies suited to diverse deployment needs.
Open access
2 source records
Adversarial Robustness in Machine Learning
Physical Unclonable Functions (PUFs) and Hardware Security
Security issues are becoming increasingly significant with the rapid evolution of Non-fungible Tokens (NFTs). As NFTs are traded as digital assets, they have emerged as prime targets for cyber attackers. In the development of NFT smart contracts, there may exist undiscovered defects that could lead to substantial financial losses if exploited. To tackle this issue, this paper presents a framework called NATLM(NFT Assistant LLM), designed to detect potential defects in NFT smart contracts. The framework effectively identifies four common types of vulnerabilities in NFT smart contracts: ERC-721 Reentrancy, Public Burn, Risky Mutable Proxy, and Unlimited Minting. Relying exclusively on large language models (LLMs) for defect detection can lead to a high false-positive rate. To enhance detection performance, NATLM integrates static analysis with LLMs, specifically Gemini Pro 1.5. Initially, NATLM employs static analysis to extract structural, syntactic, and execution flow information from the code, represented through Abstract Syntax Trees (AST) and Control Flow Graphs (CFG). These extracted features are then combined with vectors of known defect examples to create a matrix for input into the knowledge base. Subsequently, the feature vectors and code vectors of the analyzed contract are compared with the contents of the knowledge base. Finally, the LLM performs deep semantic analysis to enhance detection capabilities, providing a more comprehensive and accurate identification of potential security issues. Experimental results indicate that NATLM analyzed 8,672 collected NFT smart contracts, achieving an overall precision of 87.72%, a recall of 89.58%, and an F1 score of 88.94%. The results outperform other baseline experiments, successfully identifying four common types of defects.
With the increasing security issues in blockchain, smart contract vulnerability detection has become a research focus. Existing vulnerability detection methods have their limitations: 1) Static analysis methods struggle with complex scenarios. 2) Methods based on specialized pre-trained models perform well on specific datasets but have limited generalization capabilities. In contrast, general-purpose Large Language Models (LLMs) demonstrate impressive ability in adapting to new vulnerability patterns. However, they often underperform on specific vulnerability types compared to methods based on specialized pre-trained models. We also observe that explanations generated by generalpurpose LLMs can provide fine-grained code understanding information, contributing to improved detection performance. Inspired by these observations, we propose SAEL, a LLMbased framework for smart contract vulnerability detection. First, we design prompts targeting specific smart contract vulnerabilities to guide general-purpose LLMs in detecting vulnerabilities and providing explanations. The detection results generated by LLMs serve as prediction features. Then, we employ prompt-tuning on CodeT5 and T5 respectively to process contract code and explanations, enhancing model performance on specific tasks. To leverage the strengths of each component, we introduce Adaptive Mixture-of-Experts, a dynamic architecture for smart contract vulnerability detection. This mechanism dynamically adjusts feature weights through a Gating Network, which selects the most relevant features by applying TopK filtering and Softmax normalization, and a Multi-Head Self-Attention mechanism, which enhances cross-feature relationships by processing multiple attention heads in parallel. This design ensures that prediction results for LLMs, explanation features, and contract code features are effectively integrated through gradient optimization. The loss function focuses on the independent prediction performance of each feature and the overall performance of weighted predictions. Experimental results show that SAEL outperforms existing methods in detecting various vulnerabilities.
The results of this study highlight the effectiveness of the proposed semantic security detection framework, SSB, in identifying a wide range of vulnerabilities in smart contracts tailored for industrial control scenarios. Compared to existing tools like ZEUS, Securify, and VULTRON, SSB demonstrates superior logical coverage across various vulnerability types, as evidenced by its performance on smart contract samples. This suggests that semantic-based approaches, which integrate domain-specific invariants and runtime monitoring, can address the unique challenges of ICS, such as real-time constraints and semantic consistency between code and physical control logic. The framework's ability to model industrial invariants-covering security, functionality, consistency, time-related, and resource consumption aspects-provides a robust mechanism to prevent critical errors like unauthorized access or premature equipment operation. However, the lack of real-world ICS validation due to confidentiality constraints limits the generalizability of these findings. Future research should focus on adapting SSB for real industrial deployments, exploring scalability across diverse ICS architectures, and integrating advanced AI techniques for dynamic invariant adjustment. Additionally, addressing cross-chain interoperability and privacy concerns could further enhance the framework's applicability in complex industrial ecosystems.
Zero-Knowledge Proofs (ZKPs) are critical for privacy-preserving techniques and verifiable computation. Many ZKP protocols rely on key kernels such as the SumCheck protocol and Merkle Tree commitments to enable their key security properties. These kernels exhibit balanced binary tree computational patterns, which enable efficient hardware acceleration. Although prior work has investigated accelerating these kernels as part of an overarching ZKP protocol, exploiting this common tree pattern remains relatively underexplored. We conduct a systematic evaluation of these tree-based workloads under different traversal strategies, analyzing performance on multi-threaded CPUs and the Multifunction Tree Unit (MTU) hardware accelerator. We introduce a hardware-friendly Hybrid Traversal for binary tree that improves parallelism and scalability while significantly reducing memory traffic on hardware. Our results show that MTU achieves up to $1478\times$ speedup over CPU at DDR-level bandwidth and that our hybrid traversal outperforms breadth-first search by up to $3\times$. These findings offer practical guidance for designing efficient hardware accelerators for ZKP workloads with binary tree structures.
Hasib Ahmed Md Khyrul Islam, Huy T. Vo, Aditya Rane
In the era of synthetic media, deepfake manipulations pose a significant threat to information integrity. To address this challenge, we propose TrustDefender, a two-stage framework comprising (i) a lightweight convolutional neural network (CNN) that detects deepfake imagery in real-time extended reality (XR) streams, and (ii) an integrated succinct zero-knowledge proof (ZKP) protocol that validates detection results without disclosing raw user data. Our design addresses both the computational constraints of XR platforms while adhering to the stringent privacy requirements in sensitive settings. Experimental evaluations on multiple benchmark deepfake datasets demonstrate that TrustDefender achieves 95.3% detection accuracy, coupled with efficient proof generation underpinned by rigorous cryptography, ensuring seamless integration with high-performance artificial intelligence (AI) systems. By fusing advanced computer vision models with provable security mechanisms, our work establishes a foundation for reliable AI in immersive and privacy-sensitive applications.
Open access
2 source records
Adversarial Robustness in Machine Learning
Digital Media Forensic Detection
Generative Adversarial Networks and Image Synthesis
This paper presents the design, development, and thorough evaluation of a novel network security prototype that integrates Artificial Intelligence (AI) and blockchain technology to significantly enhance cyber security. As AI becomes increasingly embedded in cybersecurity solutions, ensuring the provenance, accountability, and integrity of AI-generated decisions has emerged as a critical challenge. Without reliable logging mechanisms, AI models remain vulnerable to adversarial manipulation and pose significant risks to critical security infrastructure. To address this, our research combines a state-of-the-art Convolutional Neural Network (CNN)-based threat detection module with a permissioned Ethereum-compatible blockchain. A custom-designed Solidity smart contract ensures secure, structured storage of comprehensive AI model metadata, while interactions with the blockchain are seamlessly managed through a lightweight Flask-based REST API. Each recorded transaction generates a unique cryptographic fingerprint, providing robust evidence for audits and forensic analyses. We evaluated the system's effectiveness through rigorous experimentation on a controlled test network, confirming immutability, traceability, and verifiable integrity of all logged metadata entries. Results demonstrated significant improvements in anomaly detection accuracy, reduced false-positive rates, and ensured real-time responsiveness essential for effective intrusion prevention. Despite controlled-environment limitations, such as transaction latency and blockchain-related operational costs, our prototype successfully establishes proof-of-concept for leveraging blockchain as an immutable audit trail for AI-driven cybersecurity systems. Future research directions include integrating advanced scaling techniques, such as layer 2 solutions, and extending the blockchain logging capabilities to cover the entire AI model lifecycle, including detailed training logs and comprehensive version histories. This work provides foundational contributions towards building trusted, auditable, and transparent AI solutions in regulated cyber security domains.
Daniel Commey, Benjamin Appiah, Griffith Selorm Klogo, Garth V. Crosby
Federated Learning (FL) enables collaborative model training on decentralized data without exposing raw data. However, the evaluation phase in FL may leak sensitive information through shared performance metrics. In this paper, we propose a novel protocol that incorporates Zero-Knowledge Proofs (ZKPs) to enable privacy-preserving and verifiable evaluation for FL. Instead of revealing raw loss values, clients generate a succinct proof asserting that their local loss is below a predefined threshold. Our approach is implemented without reliance on external APIs, using self-contained modules for federated learning simulation, ZKP circuit design, and experimental evaluation on both the MNIST and Human Activity Recognition (HAR) datasets. We focus on a threshold-based proof for a simple Convolutional Neural Network (CNN) model (for MNIST) and a multi-layer perceptron (MLP) model (for HAR), and evaluate the approach in terms of computational overhead, communication cost, and verifiability.
Smart contract vulnerabilities have led to billions in losses, yet finding actionable exploits remains challenging. Traditional fuzzers rely on rigid heuristics and struggle with complex attacks, while human auditors are thorough but slow and don't scale. Large Language Models offer a promising middle ground, combining human-like reasoning with machine speed. Early studies show that simply prompting LLMs generates unverified vulnerability speculations with high false positive rates. To address this, we present A1, an agentic system that transforms any LLM into an end-to-end exploit generator. A1 provides agents with six domain-specific tools for autonomous vulnerability discovery, from understanding contract behavior to testing strategies on real blockchain states. All outputs are concretely validated through execution, ensuring only profitable proof-of-concept exploits are reported. We evaluate A1 across 36 real-world vulnerable contracts on Ethereum and Binance Smart Chain. A1 achieves a 63% success rate on the VERITE benchmark. Across all successful cases, A1 extracts up to \$8.59 million per exploit and \$9.33 million total. Using Monte Carlo analysis of historical attacks, we demonstrate that immediate vulnerability detection yields 86-89% success probability, dropping to 6-21% with week-long delays. Our economic analysis reveals a troubling asymmetry: attackers achieve profitability at \$6,000 exploit values while defenders require \$60,000 -- raising fundamental questions about whether AI agents inevitably favor exploitation over defense.
WebAssembly has become the preferred smart contract format for various blockchain platforms due to its high portability and near-native execution speed. To effectively understand WebAssembly contracts, it is crucial to recover high-level type signatures because of the limited type information that WebAssembly provides. However, existing studies on type inference for smart contracts primarily center around Ethereum Virtual Machine bytecode, which is not applicable to WebAssembly owing to their differing targets and runtime semantics. This paper introduces WasmHint, a novel solution that leverages deep learning inference to automatically recover high-level parameter and return types from WebAssembly contracts. More specifically, WasmHint constructs a wCFG representation to clarify dependencies within WebAssembly code and simulates its execution to capture type-related operational information. By learning comprehensive code semantics, it infers parameter and return types, with a semantic corrector designed to enhance information coordination. We conduct experiments on a newly constructed dataset containing 77,208 WebAssembly contract functions. The results demonstrate that WasmHint achieves inference accuracies of 80.0% for parameter types and 95.8% for return types, with average improvements of 86.6% and 34.0% over the baseline methods, respectively.
The rapid evolution of phishing attacks targeting email, chat, and social media platforms poses a significant threat to digital security, with a reported 667% surge in spear-phishing during the 2020 COVID-19 crisis [1]. Current AI-based detection systems face challenges in dataset diversity, adversarial robustness, computational scalability, model interpretability, and privacy preservation, limiting their efficacy in real-time, multi-platform environments. This paper introduces PhishGuard, an innovative framework for real-time phishing detection, designed to overcome these limitations. PhishGuard integrates lightweight transformer models (e.g., distilled BERT), hybrid detection techniques combining natural language processing (NLP), propagation analysis, and user behavior analysis, and explainable AI (XAI) methods like SHAP and LIME for transparent decision-making. Privacy-preserving techniques, including federated learning and local differential privacy, ensure secure processing of sensitive user data. Evaluated on diverse datasets such as PhiKitA, Enron, and a custom social media corpus, PhishGuard achieves up to 97.5% accuracy, 94% F1-score, and inference times below 5 ms, demonstrating scalability for resource-constrained devices. The framework also incorporates zero-knowledge proofs for verifiable inference, addressing trust and integrity concerns. By tackling cross-domain generalization, adversarial robustness, and real-time performance, PhishGuard offers a scalable, user centric solution for secure digital communications, with applications in finance, healthcare, and social media platforms. Future enhancements include multilingual support and image based phishing detection, paving the way for a comprehensive defense against evolving cyber threats.
Ruichao Liang, Jing Chen, Ruochen Cao, Kun He · 8 authors
Smart contracts, as Turing-complete programs managing billions of assets in decentralized finance, are prime targets for attackers. While fuzz testing seems effective for detecting vulnerabilities in these programs, we identify several significant challenges when targeting smart contracts: (i) the stateful nature of these contracts requires stateful exploration, but current fuzzers rely on transaction sequences to manipulate contract states, making the process inefficient; (ii) contract execution is influenced by the continuously changing blockchain environment, yet current fuzzers are limited to local deployments, failing to test contracts in real-world scenarios. These challenges hinder current fuzzers from uncovering hidden vulnerabilities, i.e., those concealed in deep contract states and specific blockchain environments. In this paper, we present SmartShot, a mutable snapshot-based fuzzer to hunt hidden vulnerabilities within smart contracts. We innovatively formulate contract states and blockchain environments as directly fuzzable elements and design mutable snapshots to quickly restore and mutate these elements. SmartShot features a symbolic taint analysis-based mutation strategy along with double validation to soundly guide the state mutation. SmartShot mutates blockchain environments using contract’s historical on-chain states, providing real-world execution contexts. We propose a snapshot checkpoint mechanism to integrate mutable snapshots into SmartShot’s fuzzing loops. These innovations enable SmartShot to effectively fuzz contract states, test contracts across varied and realistic blockchain environments, and support on-chain fuzzing. Experimental results show that SmartShot is effective to detect hidden vulnerabilities with the highest code coverage and lowest false positive rate. SmartShot is 4.8× to 20.2× faster than state-of-the-art tools, identifying 2,150 vulnerable contracts out of 42,738 real-world contracts which is 2.1× to 13.7× more than other tools. SmartShot has demonstrated its real-world impact by detecting vulnerabilities that are only discoverable on-chain and uncovering 24 0-day vulnerabilities in the latest 10,000 deployed contracts.
Federated Learning (FL) has emerged as a critical paradigm for enabling privacy-preserving machine learning, particularly in regulated sectors such as finance and healthcare. However, standard FL strategies often encounter significant operational challenges related to fault tolerance, system resilience against concurrent client and server failures, and the provision of robust, verifiable privacy guarantees essential for handling sensitive data. These deficiencies can lead to training disruptions, data loss, compromised model integrity, and non-compliance with data protection regulations (e.g., GDPR, CCPA). This paper introduces Differentially Private Resilient Temporal Federated Learning (DP-RTFL), an advanced FL framework designed to ensure training continuity, precise state recovery, and strong data privacy. DP-RTFL integrates local Differential Privacy (LDP) at the client level with resilient temporal state management and integrity verification mechanisms, such as hash-based commitments (referred to as Zero-Knowledge Integrity Proofs or ZKIPs in this context). The framework is particularly suited for critical applications like credit risk assessment using sensitive financial data, aiming to be operationally robust, auditable, and scalable for enterprise AI deployments. The implementation of the DP-RTFL framework is available as open-source.
Filippo Scaramuzza, Giovanni Quattrocchi, Damian A. Tamburri
As Artificial Intelligence (AI) systems, particularly those based on machine learning (ML), become integral to high-stakes applications, their probabilistic and opaque nature poses significant challenges to traditional verification and validation methods. These challenges are exacerbated in regulated sectors requiring tamper-proof, auditable evidence, as highlighted by apposite legal frameworks, e.g., the EU AI Act. Conversely, Zero-Knowledge Proofs (ZKPs) offer a cryptographic solution that enables provers to demonstrate, through verified computations, adherence to set requirements without revealing sensitive model details or data. Through a systematic survey of ZKP protocols, we identify five key properties (non-interactivity, transparent setup, standard representations, succinctness, and post-quantum security) critical for their application in AI validation and verification pipelines. Subsequently, we perform a follow-up systematic survey analyzing ZKP-enhanced ML applications across an adaptation of the Team Data Science Process (TDSP) model (Data & Preprocessing, Training & Offline Metrics, Inference, and Online Metrics), detailing verification objectives, ML models, and adopted protocols. Our findings indicate that current research on ZKP-Enhanced ML primarily focuses on inference verification, while the data preprocessing and training stages remain underexplored. Most notably, our analysis identifies a significant convergence within the research domain toward the development of a unified Zero-Knowledge Machine Learning Operations (ZKMLOps) framework. This emerging framework leverages ZKPs to provide robust cryptographic guarantees of correctness, integrity, and privacy, thereby promoting enhanced accountability, transparency, and compliance with Trustworthy AI principles.
Large Language Models (LLMs) have shown great promise in code analysis and auditing; however, they still struggle with hallucinations and limited context-aware reasoning. We introduce SmartAuditFlow, a novel Plan-Execute framework that enhances smart contract security analysis through dynamic audit planning and structured execution. Unlike conventional LLM-based auditing approaches that follow fixed workflows and predefined steps, SmartAuditFlow dynamically generates and refines audit plans based on the unique characteristics of each smart contract. It continuously adjusts its auditing strategy in response to intermediate LLM outputs and newly detected vulnerabilities, ensuring a more adaptive and precise security assessment. The framework then executes these plans step by step, applying a structured reasoning process to enhance vulnerability detection accuracy while minimizing hallucinations and false positives. To further improve audit precision, SmartAuditFlow integrates iterative prompt optimization and external knowledge sources, such as static analysis tools and Retrieval-Augmented Generation (RAG). This ensures audit decisions are contextually informed and backed by real-world security knowledge, producing comprehensive security reports. Extensive evaluations across multiple benchmarks demonstrate that SmartAuditFlow outperforms existing methods, achieving 100 percent accuracy on common and critical vulnerabilities, 41.2 percent accuracy for comprehensive coverage of known smart contract weaknesses in real-world projects, and successfully identifying all 13 tested CVEs. These results highlight SmartAuditFlow's scalability, cost-effectiveness, and superior adaptability over traditional static analysis tools and contemporary LLM-based approaches, establishing it as a robust solution for automated smart contract auditing.
With the rise of machine learning techniques, ensuring the fairness of decisions made by machine learning algorithms has become of great importance in critical applications. However, measuring fairness often requires full access to the model parameters, which compromises the confidentiality of the models. In this paper, we propose a solution using zero-knowledge proofs, which allows the model owner to convince the public that a machine learning model is fair while preserving the secrecy of the model. To circumvent the efficiency barrier of naively proving machine learning inferences in zero-knowledge, our key innovation is a new approach to measure fairness only with model parameters and some aggregated information of the input, but not on any specific dataset. To achieve this goal, we derive new bounds for the fairness of logistic regression and deep neural network models that are tighter and better reflecting the fairness compared to prior work. Moreover, we develop efficient zero-knowledge proof protocols for common computations involved in measuring fairness, including the spectral norm of matrices, maximum, absolute value, and fixed-point arithmetic. We have fully implemented our system, FairZK, that proves machine learning fairness in zero-knowledge. Experimental results show that FairZK is significantly faster than the naive approach and an existing scheme that use zero-knowledge inferences as a subroutine. The prover time is improved by 3.1x--1789x depending on the size of the model and the dataset. FairZK can scale to a large model with 47 million parameters for the first time, and generates a proof for its fairness in 343 seconds. This is estimated to be 4 orders of magnitude faster than existing schemes, which only scale to small models with hundreds to thousands of parameters.