In the past few years, several attacks against the vulnerabilities of EOSIO smart contracts have caused severe financial losses to this prevalent blockchain platform. As a lightweight test-generation approach, grey-box fuzzing can open up the possibility of improving the security of EOSIO smart contracts. However, developing a practical grey-box fuzzer for EOSIO smart contracts from scratch is time-consuming and requires a deep understanding of EOSIO internals. In this work, we proposed AntFuzzer, the first highly extensible grey-box fuzzing framework for EOSIO smart contracts. AntFuzzer implements a novel approach that interfaces AFL to conduct AFL-style grey-box fuzzing on EOSIO smart contracts. Compared to black-box fuzzing tools, AntFuzzer can effectively trigger those hard-to-cover branches. It achieved an improvement in code coverage on 37.5% of smart contracts in our benchmark dataset. AntFuzzer provides unified interfaces for users to easily develop new detection plugins for continually emerging vulnerabilities. We have implemented 6 detection plugins on AntFuzzer to detect major vulnerabilities of EOSIO smart contracts. In our large-scale fuzzing experiments on 4,616 real-world smart contracts, AntFuzzer successfully detected 741 vulnerabilities. The results demonstrate the effectiveness and efficiency of AntFuzzer and our detection pl
Federated learning is a type of privacy-preserving, collaborative machine learning. Instead of sharing raw data, the federated learning process cooperatively exchanges the model parameters and aggregates them in a decentralized manner through multiple users. In this study, we designed and implemented a hierarchical blockchain system using a public blockchain for a federated learning process without a trusted curator. This prevents model-poisoning attacks and provides secure updates of a global model. We conducted a comprehensive empirical study to characterize the performance of federated learning in our testbed and identify potential performance bottlenecks, thereby gaining a better understanding of the system.
Smart contracts have been widely and rapidly used to automate financial and business transactions together with blockchains, helping people make agreements while minimizing trusts. With millions of smart contracts deployed on blockchain, various bugs and vulnerabilities in smart contracts have emerged. Following the rapid development of deep learning, many recent studies have used deep learning for vulnerability detection to conduct security checks before deploying smart contracts. These approaches show effective results on detecting whether a smart contract is vulnerable or not whereas their results on locating suspicious statements responsible for the detected vulnerability are still unsatisfactory.
As computer programs run on top of blockchain, smart contracts have proliferated a myriad of decentralized applications while bringing security vulnerabilities, which may cause huge financial losses. Thus, it is crucial and urgent to detect the vulnerabilities of smart contracts. However, existing fuzzers for smart contracts are still inefficient to detect sophisticated vulnerabilities that require specific vulnerable transaction sequences to trigger. To address this challenge, we propose a novel vulnerability-guided fuzzer based on reinforcement learning, namely RLF, for generating vulnerable transaction sequences to detect such sophisticated vulnerabilities in smart contracts. In particular, we firstly model the process of fuzzing smart contracts as a Markov decision process to construct our reinforcement learning framework. We then creatively design an appropriate reward with consideration of both vulnerability and code coverage so that it can effectively guide our fuzzer to generate specific transaction sequences to reveal vulnerabilities, especially for the vulnerabilities related to multiple functions. We conduct extensive experiments to evaluate RLF’s performance. The experimental results demonstrate that our RLF outperforms state-of-the-art vulnerability-detection tools (e.g., detecting 8%-69% more vulnerabilities within 30 minutes).
When do classical zero-knowledge protocols remain secure against quantum attacks? In this work, we develop the techniques, tools, and abstractions necessary to answer this question for foundational protocols:1)We prove that the Goldreich-Micali-Wigderson protocol for graph non-isomorphism and the Feige-Shamir protocol for NP remain zero-knowledge against quantum adversaries. At the heart of our proof is a new quantum rewinding technique that enables extracting information from multiple invocations of a quantum adversary without disturbing its state.2)We prove that the Goldreich-Kahan protocol for NP is post-quantum zero knowledge using a simulator that can be seen as a natural quantum extension of the classical simulator.Our results achieve negligible simulation error, appearing to contradict a recent impossibility result due to Chia-Chung-Liu-Yamakawa (FOCS 2021). This brings us to our final contribution:3.We introduce coherent-runtime expected quantum polynomial time, a simulation notion that (a) precisely captures all of our zero-knowledge simulators, (b) cannot break any polynomial hardness assumptions, (c) implies strict polynomial-time ε-simulation and (d) is not subject to the CCLY impossibility. In light of our positive results and the CCLY negative results, we propose coherent-runtime simulation to be the appropriate quantum analogue of classical expected polynomial-time simulation.
Shuhao Zheng, Yanxi Lin, Yang Yu, Ye Yuan · 6 authors
Auditing the information leakage of latent sensitive features during the transborder data flow has attracted sufficient attention from global digital regulators. However, there is missing a technical approach for the audit practice due to two technical challenges. Firstly, there is a lack of theory and tools for measuring the information of sensitive latent features in a dataset. Secondly, the transborder data flow involves multi-stakeholders with diverse interests, which means the audit must be trustless. Despite the tremendous efforts in protecting data privacy, an important issue that has long been neglected is that the transmitted data in data flows can leak other regulated information that is not explicitly contained in the data, leading to unaware information leakage risks. To unveil such risks trustfully before the actual data transfer, we propose FIAT, a Fine-grained Information Audit system for Trustless transborder data flow. In FIAT, we use a learning approach to quantify the amount of information leakage, while the technologies of zero-knowledge proof and smart contracts are applied to provide trustworthy and privacy-preserving auditing results. Experiments show that large information leakage can boost the predictability of uninvolved information using simple machine-learning models, revealing the importance of information auditing. Further performance benchmarking also validates the efficiency and scalability of the FIAT auditing system.
With the wide application and development of blockchain technology, the past years have witnessed the emergence of various cybercrimes, which have caused a huge amount of economic loss. Among them, phishing scams on the blockchain are regarded as a serious threat to the trading security of the blockchain ecosystem. By modeling the transaction data of blockchain as a network, a series of graph-based phishing detection frameworks have been proposed. Enlightened by adversarial attacks of graph data, we propose to verify the robustness of current phishing detection frameworks under intentional attackers aiming to hide phishing behaviors. In this study, we first propose a general phishing detection framework based on feature engineering and then propose a phishing hiding framework combing the greedy selection mechanism with four phishing hiding strategies to measure the robustness of the proposed general detection models. Extensive experiments evaluate the detective performance of the phishing detection model and its robustness against the hiding framework. The experimental results indicate that the detective model based on feature engineering is rather fragile under adversarial attacks.
Abstract The realization of fully autonomous vehicles in great scale is already in sight. However, to be safe and reliable, autonomous driving requires a significant amount of communication with cloud services, which are prone to many kinds of failures and malicious attacks. This paper introduces the use of AIngle, a novel distributed-ledger technology developed to guarantee real-time encrypted communication among IoT devices. Not only AIngle tackles various kinds of malicious attacks by design, such as Sybil attacks, but it also enables semantic distributed applications (Sem D Apps). Sem D Apps decentralize supporting driving services, such as traffic and route mapping, which are typically centralized in cloud services. AIngle provides a new paradigm for Sem D Apps, that will support future-proofed, fast, reliable, intelligent distributed applications for the autonomous vehicle industry.
Zhendong Wu, Shan Li, Bin Wang, Tianjian Liu · 7 authors
Decentralized applications and smart contracts running on Ethereum have been widely applied in many fields, such as finance and logistics. On the one hand, the number of smart contracts surges. On the other hand, smart contracts often carry million dollars. Once there is a severe vulnerability in a smart contract, huge losses may occur. Detection tools like Oyente and Mythril were developed for vulnerabilities in smart contracts. However, many of these tools many detection tools do not use intrinsic features in the contract, resulting in ineffectiveness in the detection. In addition, many tools rely on symbolic execution and lack automation in the detection, resulting in inefficiency and in incapacity in vetting huge number of emerging smart contract on Ethereum. In our work, we are motivated to enhance the effectiveness and efficiency of vulnerability detection in contracts. First, we use opcodes as static features and use bigram to build the opcode feature space that is further optimized with Mtfidf. Second, We use deep learning algorithms namely, CNN, LSTM, CNN-BiLSTM, and ResNets, for the smart contract vulnerability detection. Extensive experimental results based on real-world smart contracts show that our methods are promising. Mtfidf significantly improves the detection performance. The best detection performance with ResNets reaches 82% in terms of Macro-F1.
WebAssembly (Wasm) smart contracts have shown growing popularity across blockchains (e.g., EOSIO) recently. Similar to Ethereum smart contracts, Wasm smart contracts suffer from various attacks exploiting their vulnerabilities. Even worse, few developers released the source code of their Wasm smart contracts for security review, raising the bar for uncovering vulnerable contracts. Although a few approaches have been proposed to detect vulnerable Wasm smart contracts, they have several major limitations, e.g., low code coverage, low accuracy and lack of scalability, unable to produce exploit payloads, etc. To fill the gap, in this paper, we design and develop WASAI, a new concolic fuzzer for uncovering vulnerabilities in Wasm smart contract after tackling several challenging issues. We conduct extensive experiments to evaluate WASAI, and the results show that it outperforms the state-of-the-art methods. For example, it achieves 2x code coverage than the baselines and surpasses them in detection accuracy, with an F1-measure of 99.2%. Moreover, WASAI can handle complicated contracts (e.g., contracts with obfuscation and sophisticated verification). Applying WASAI to 991 deployed smart contracts in the wild, we find that over 70% of smart contracts are vulnerable. By the time of this study, over 300 vulnerable contracts have not been patched and are still operating on the EOSIO Mainnet. One fake EOS vulnerability reported to the EOSIO ecosystem was recently assigned a CVE identifier (CVE-2022-27134).
Privacy-preserving federated learning allows multiple users to jointly train a model with coordination of a central server. The server only learns the final aggregation result, thereby preventing leakage of the users’ (private) training data from the individual model updates. However, keeping the individual updates private allows malicious users to degrade the model accuracy without being detected, also known as Byzantine attacks. Best existing defenses against Byzantine workers rely on robust rank-based statistics, e.g., setting robust bounds via the median of updates, to find malicious updates. However, implementing privacy-preserving rank-based statistics, especially median-based, is nontrivial and unscalable in the secure domain, as it requires sorting of all individual updates. We establish the first private robustness check that uses high break point rank-based statistics on aggregated model updates. By exploiting randomized clustering, we significantly improve the scalability of our defense without compromising privacy. We leverage the derived statistical bounds in zero-knowledge proofs to detect and remove malicious updates without revealing the private user updates. Our novel framework, zPROBE, enables Byzantine resilient and secure federated learning. We show the effectiveness of zPROBE on several computer vision benchmarks. Empirical evaluations demonstrate that zPROBE provides a low overhead solution to defend against state-of-the-art Byzantine attacks while preserving privacy.
Ethereum smart contracts may incur security vulnerabilities. Fuzzing is an industry-standard practice to detect them in improving the dependability of programs. Existing fuzz testing techniques for Ethereum smart contracts are insensitive to whether consecutive seeds of the same function are used for fuzzing the smart contract under test. Nonetheless, smart contracts are often designed to have collaborations among different functions for business activity to complete. We wonder whether this mismatch will make fuzzing techniques less effective than they should be. In this paper, to the best of our knowledge, we present the first work to show that security vulnerability detection can be significantly more effective in smart contract fuzzing if the entry functions of recent past test cases can be distinct. The empirical results show that the performance boost can be as large as 10.4% by simply enabling any test case not invoking the same entry functions as a few recent past test cases. The empirical result also shows that the cost-effectiveness also increases by up to 21.9%.
Blockchain applications go far beyond cryptocurrency. As an essential blockchain tool, smart contracts are executable programs that establish an agreement between two parties. Millions of dollars of transactions attract hackers at a hastened pace, and cyber-attacks have caused large economic losses in the past. Due to this, the industry is seeking robust and effective methods to detect vulnerabilities in smart contracts to ultimately provide a remedy. The industry has been utilizing static analysis tools to reveal security gaps, which requires an understanding and insight over all possible execution paths to identify known contract vulnerabilities. Yet, the computational complexity increases as the path gets deeper. Recently, researchers have been proposing ML-driven intelligent techniques aiming to improve the efficiency and detection rate. Such solutions can provide quicker and more robust detection options than the traditionally used static analysis tools. As of this publication date, there is currently no published survey paper on smart contract vulnerability detection mechanisms using ML models. In order to set the ground for further development of ML-driven solutions, in this survey paper, we extensively reviewed and summarized a wide variety of ML-driven intelligent detection mechanism from the following databases: Google Scholar, Engineering Village, Springer, Web of Science, Academic Search Premier, and Scholars Portal Journal. In conclusion, we provided our insights on common traits, limitations and advancement of ML-driven solutions proposed for this field.
May 20, 2022·2022 3rd International Conference on Computer Vision, Image and Deep Learning & International Conference on Computer Engineering and Applications (CVIDL & ICCEA)
The security problem within cryptocurrencies has high priority since it is published. The news of cryptocurrencies attack is keeping rising, and types of attacks are recorded as the classical security issues such as the most famous one decentralized autonomous organization (DAO) attack and which caused 3.6 million cryptocurrencies stolen. Thus, with the ripe fuzzing technique, using such testing skills on smart contracts to discover more bugs and leaks hidden in cryptocurrencies in a more automatic and intelligent approach could bring huge benefits. This paper is aimed to analyze the variation from the very first smart contract tools to the newest one, to discuss their pros and cons of them. This experiment mainly chose three fuzzing tools, several basic and classic smart contracts as tested material, and recorded the performance of fuzzing contracts of each tool. And the results are clear to show that even the newest fuzzer has a better performance than the other, but it is better to use it depending on the specific requirements while the rest of the two fuzzing tools has some advantages in some aspect. Therefore, this paper could make guidance for the later researcher to use exactly the type of fuzzer they need.
Privacy and data security have become the new hot topic for regulators in recent years. As a result, Federated Learning (FL) (also called collaborative learning) has emerged as a new training paradigm that allows multiple, geographically distributed nodes to learn a Deep Learning (DL) model together without sharing their data. Blockchain is becoming a new trend as data protection and privacy are concerns in many sectors. Technology is leading the world and transforming into a global village where everything is accessible and transparent. We have presented a blockchain enabled security model using FL that can generate an enhanced DL model without sharing data and improve privacy through higher security and access rights to data. However, existing FL approaches also have unique security vulnerabilities that malicious actors can exploit and compromise the trained model. The FL method is compared to the other known approaches. Users are more likely to choose the latter option, i.e., providing local but private data to the server and using ML apps, performing ML operations on the devices without benefiting from other users’ data, and preventing direct access to raw data and local training of ML models. FL protects data privacy and reduces data transfer overhead by storing raw data on devices and combining locally computed model updates. We have investigated the feasibility of data and model poisoning attacks under a blockchain-enabled FL system built alongside the Ethereum network and the traditional FL system (without blockchain). This work fills a knowledge gap by proposing a transparent incentive mechanism that can encourage good behavior among participating decentralized nodes and avoid common problems and provides knowledge for the FL security literature by investigating current FL systems.
Shunhui Ji, Shaoqing Zhu, Pengcheng Zhang, Hai Dong · 5 authors
Smart contracts are commonly deployed for safety-critical applications, the quality assurance of which has been a vital factor. Test cases are standard means to ensure the correctness of data flows in smart contracts. To more efficiently generate test cases with high coverage, we propose an improved genetic algorithm-based test-case generation approach for smart contract data flow testing. Our approach introduces the theory of particle swarm optimization into the genetic algorithm, which reduces the influence brought by the randomness of genetic operations and enhances its capability to find global optima. A set of 30 real smart contracts deployed on Ethereum and GitHub is collected to perform the experimental study, on which our approach is compared with three baseline approaches. The experimental results show that, in most cases, the coverage of the test cases generated by our approach is significantly higher than the baseline approaches with relatively lower numbers of iterations and lower execution time.
Smart contracts are self-executing programs that are stored on the blockchain. Once a smart contract is compiled and deployed on the blockchain, it cannot be modified. Therefore, having a bug-free smart contract is vital. To ensure a bug-free smart contract, it must be tested thoroughly. However, little is known about how developers test smart contracts in practice. Our study explores 139 open source smart contract projects that are written in Solidity to investigate the state of smart contract testing from three dimensions: (1) the developers working on the tests, (2) the used testing frameworks and testnets and (3) the type of tests that are conducted. We found that mostly core developers of a project are responsible for testing the contracts. Second, developers typically use only functional testing frameworks to test a smart contract, with Truffle being the most popular one. Finally, our results show that functional testing is conducted in most of the studied projects (93%), security testing is only performed in a few projects (9.4%) and traditional performance testing is conducted in none. In addition, we found 34 projects that mentioned or published external audit reports.
Artificial intelligence (AI) and blockchain are two of the most disruptive technologies in recent years. Blockchain is widely regarded as a trust machine because of its decentralization, non-tampering, anonymity and traceability. AI provides machines with cognitive functions, including learning, reasoning, and adaptation based on the collected data, which enables human-like machines possess intelligence and decision-making capabilities. Also, both technologies are data-driven, and thus there are rapidly growing interests in integrating them for trustworthy artificial intelligence and intelligent blockchain. In this paper, we review the related research on the integration of AI and blockchain, mainly analyzing how blockchain technology can improve AI from five aspects and pointing out the future research direction of these two technologies. And our research shows that blockchain can drive various components of AI including data, algorithms, and computational power to higher levels.
Alex Groce, Kush Jain, Rijnard van Tonder, Goutamkumar Tulajappa Kalburgi · 5 authors
Bitcoin is one of the most prominent distributed software systems in the world. This paper describes an effort to investigate and enhance the effectiveness of the Bitcoin Core fuzzing effort. The effort initially began as a query about how to escape saturation in the fuzzing effort, but developed into a more general exploration. This paper summarizes the outcomes of a two-week focused effort. While the effort found no smoking guns indicating major test/fuzz weaknesses, it produced a large number of additional fuzz corpus entries, increased the set of fuzzers used for Bitcoin Core, and ran mutation analysis of Bitcoin Core fuzz targets, with a comparison to Bitcoin functional tests and other cryptocurrencies’ tests. Our conclusion is that for high quality fuzzing efforts, improvements to the oracle may be the best way to get more out of fuzzing.
As one of the most complex types of vulnerabilities, reentrancy poses a significant threat to smart contract development. Indeed, millions of dollars have evaporated due to reentrancy vulnerabilities of smart contracts in past years. In this article, we propose a new approach to detect reentrancy vulnerabilities using fuzz testing and develop a novel tool named ReDefender. Our approach consists of three main steps: 1)preprocess contract to be detected:when a contract is uploaded, its source code will be preprocessed to extract candidate pool for fuzzing and dependency graph which guides the automatic deployment of contracts; 2)fuzzing input generation:fuzzing input is generated to constitute transactions which will be sent to an agent contract to stimulate attacks, where runtime information is collected and recorded in the execution log during each execution; and 3)vulnerability verification:the execution log is analyzed to determine whether a reentrancy process occurs and whether the reentrancy process is malicious. We conduct comparative experiments on 204 tagged smart contracts and 90 injected contracts. The results show higher accuracy and lower false negative rate of ReDefender than that of the other three famous tools. Moreover, we conduct an experiment on 4776 real-world contracts demonstrating the ability of ReDefender to find reentrancy vulnerabilities that really cause economic losses.
Francesco Zola, Lander Segurola-Gil, Jan L. Bruse, Mikel Galar · 5 authors
Abstract Classification of Bitcoin entities is an important task to help Law Enforcement Agencies reduce anonymity in the Bitcoin blockchain network and to detect classes more tied to illegal activities. However, this task is strongly conditioned by a severe class imbalance in Bitcoin datasets. Existing approaches for addressing the class imbalance problem can be improved considering generative adversarial networks (GANs) that can boost data diversity. However, GANs are mainly applied in computer vision and natural language processing tasks, but not in Bitcoin entity behaviour classification where they may be useful for learning and generating synthetic behaviours. Therefore, in this work, we present a novel approach to address the class imbalance in Bitcoin entity classification by applying GANs. In particular, three GAN architectures were implemented and compared in order to find the most suitable architecture for generating Bitcoin entity behaviours. More specifically, GANs were used to address the Bitcoin imbalance problem by generating synthetic data of the less represented classes before training the final entity classifier. The results were used to evaluate the capabilities of the different GAN architectures in terms of training time, performance, repeatability, and computational costs. Finally, the results achieved by the proposed GAN-based resampling were compared with those obtained using five well-known data-level preprocessing techniques. Models trained with data resampled with our GAN-based approach achieved the highest accuracy improvements and were among the best in terms of precision, recall and f1-score. Together with Random Oversampling (ROS), GANs proved to be strong contenders in addressing Bitcoin class imbalance and consequently in reducing Bitcoin entity anonymity (overall and per-class classification performance). To the best of our knowledge, this is the first work to explore the advantages and limitations of GANs in generating specific Bitcoin data and “attacking” Bitcoin anonymity. The proposed methods ultimately demonstrate that in Bitcoin applications, GANs are indeed able to learn the data distribution and generate new samples starting from a very limited class representation, which leads to better detection of classes related to illegal activities.
José L. Muñoz, Marta Bellés, Miguel Isabel, Albert Rubio · 5 authors
A zero-knowledge (ZK) proof guarantees that the result of a computation is correct while keeping part of the computation details private. Some ZK proofs are tiny and can be verified in short time, which makes them one of the most promising technologies for solving two key aspects: the challenge of enabling privacy to public and transparent distributed ledgers and, enhancing the scalability limitations of distributed ledgers. Most practical ZK systems require the computation to be expressed as an arithmetic circuit that is encoded as a set of equations called rank-1 constraint system (R1CS). In this paper, we present \circom, a programming language and a compiler for designing arithmetic circuits that are compiled to R1CS. More precisely, with \circom, programmers can design arithmetic circuits, and the compiler outputs (i) a file with the R1CS description, (ii) \wasm and \cpp programs to efficiently compute all values of the circuit. We also provide an open-source library called \circomlib, with multiple circuit templates. Moreover, \circom can be complemented with \snarkjs, a tool for generating and validating ZK proofs from R1CS. Altogether, our software tools abstract the complexity of the proving mechanisms and provide a friendly interface to model low-level descriptions of arithmetic circuits.
The Open Vote Network is a self-tallying decentralized e-voting protocol suitable for boardroom elections. Currently, it has two Ethereum-based implementations: the first, by McCorry et al., has a scalability issue since all the computations are performed on-chain. The second implementation, by Seifelnasr et al., solves this issue partially by assigning a part of the heavy computations to an off-chain untrusted administrator in a verifiable manner. As a side effect, this second implementation became not dispute-free; there is a need for a tally dispute phase where an observer interrupts the protocol when the administrator cheats, i.e., announces a wrong tally result. In this work, we propose a new smart contract design to tackle the problems in the previous implementations by (i) preforming all the heavy computations off-chain hence achieving higher scalability, and (ii) utilizing zero-knowledge Succinct Non-interactive Argument of Knowledge (zk-SNARK) to verify the correctness of the off-chain computations, hence maintaining the dispute-free property. To demonstrate the effectiveness of our design, we develop prototype implementations on Ethereum and conduct multiple experiments for different implementation options that show a trade-off between the zk-SNARK proof generation time and the smart contract gas cost, including an implementation in which the smart contract consumes a constant amount of gas independent of the number of voters.
Federated learning is known to be vulnerable to both security and privacy issues. Existing research has focused either on preventing poisoning attacks from users or on concealing the local model updates from the server, but not both. However, integrating these two lines of research remains a crucial challenge since they often conflict with one another with respect to the threat model. In this work, we develop a principle framework that offers both privacy guarantees for users and detection against poisoning attacks from them. With a new threat model that includes both an honest-but-curious server and malicious users, we first propose a secure aggregation protocol using homomorphic encryption for the server to combine local model updates in a private manner. Then, a zero-knowledge proof protocol is leveraged to shift the task of detecting attacks in the local models from the server to the users. The key observation here is that the server no longer needs access to the local models for attack detection. Therefore, our framework enables the central server to identify poisoned model updates without violating the privacy guarantees of secure aggregation.