Yinxing Xue

Jie Li, Yi Liu, Chongyang Liu et al.

Large Language Models (LLMs) have become increasingly popular for their advanced text generation capabilities across various domains. However, like any software, they face security challenges, including the risk of 'jailbreak' attacks that manipulate LLMs to produce prohibited content. A particularly underexplored area is the Multilingual Jailbreak attack, where malicious questions are translated into various languages to evade safety filters. Currently, there is a lack of comprehensive empirical studies addressing this specific threat. To address this research gap, we conducted an extensive empirical study on Multilingual Jailbreak attacks. We developed a novel semantic-preserving algorithm to create a multilingual jailbreak dataset and conducted an exhaustive evaluation on both widely-used open-source and commercial LLMs, including GPT-4 and LLaMa. Additionally, we performed interpretability analysis to uncover patterns in Multilingual Jailbreak attacks and implemented a fine-tuning mitigation method. Our findings reveal that our mitigation strategy significantly enhances model defense, reducing the attack success rate by 96.2%. This study provides valuable insights into understanding and mitigating Multilingual Jailbreak attacks.

Jie Li, Yi Liu, Chongyang Liu et al.

Large Language Models (LLMs) like OpenAI's GPT series, Anthropic's Claude, and Meta's LLaMa have shown remarkable capabilities in text generation. However, their susceptibility to toxic prompts presents significant security challenges. This paper investigates alignment techniques, including Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF), to mitigate these risks. We conduct an empirical study on refusal patterns across nine LLMs, revealing that models with uniform refusal patterns, such as Claude3, exhibit higher security. Based on these findings, we propose self-distilling and cross-model distilling methods to enhance LLM security. Our results show that these methods significantly improve refusal rates and reduce unsafe content, with cross-model distilling achieving refusal rates close to Claude3's 94.51%. These findings underscore the potential of distillation-based alignment in securing LLMs against toxic prompts.

Yinxing Xue, Jiaming Ye, Wei Zhang et al.

Smart contract transactions are increasingly interleaved by cross-contract calls. While many tools have been developed to identify a common set of vulnerabilities, the cross-contract vulnerability is overlooked by existing tools. Cross-contract vulnerabilities are exploitable bugs that manifest in the presence of more than two interacting contracts. Existing methods are however limited to analyze a maximum of two contracts at the same time. Detecting cross-contract vulnerabilities is highly non-trivial. With multiple interacting contracts, the search space is much larger than that of a single contract. To address this problem, we present xFuzz, a machine learning guided smart contract fuzzing framework. The machine learning models are trained with novel features (e.g., word vectors and instructions) and are used to filter likely benign program paths. Comparing with existing static tools, machine learning model is proven to be more robust, avoiding directly adopting manually-defined rules in specific tools. We compare xFuzz with three state-of-the-art tools on 7,391 contracts. xFuzz detects 18 exploitable cross-contract vulnerabilities, of which 15 vulnerabilities are exposed for the first time. Furthermore, our approach is shown to be efficient in detecting non-cross-contract vulnerabilities as well -- using less than 20% time as that of other fuzzing tools, xFuzz detects twice as many vulnerabilities.

Weiwei Zhang, Shengjian Guo, Hongyu Zhang et al.

Software clone detection identifies similar code snippets. It has been an active research topic that attracts extensive attention over the last two decades. In recent years, machine learning (ML) based detectors, especially deep learning-based ones, have demonstrated impressive capability on clone detection. It seems that this longstanding problem has already been tamed owing to the advances in ML techniques. In this work, we would like to challenge the robustness of the recent ML-based clone detectors through code semantic-preserving transformations. We first utilize fifteen simple code transformation operators combined with commonly-used heuristics (i.e., Random Search, Genetic Algorithm, and Markov Chain Monte Carlo) to perform equivalent program transformation. Furthermore, we propose a deep reinforcement learning-based sequence generation (DRLSG) strategy to effectively guide the search process of generating clones that could escape from the detection. We then evaluate the ML-based detectors with the pairs of original and generated clones. We realize our method in a framework named CloneGen. CloneGen In evaluation, we challenge the two state-of-the-art ML-based detectors and four traditional detectors with the code clones after semantic-preserving transformations via the aid of CloneGen. Surprisingly, our experiments show that, despite the notable successes achieved by existing clone detectors, the ML models inside these detectors still cannot distinguish numerous clones produced by the code transformations in CloneGen. In addition, adversarial training of ML-based clone detectors using clones generated by CloneGen can improve their robustness and accuracy. CloneGen Meanwhile, compared with the commonly-used heuristics, the DRLSG strategy has shown the best effectiveness in generating code clones to decrease the detection accuracy of the ML-based detectors.

Hongxu Chen, Shengjian Guo, Yinxing Xue et al.

Grey-box fuzz testing has revealed thousands of vulnerabilities in real-world software owing to its lightweight instrumentation, fast coverage feedback, and dynamic adjusting strategies. However, directly applying grey-box fuzzing to input-dependent multithreaded programs can be extremely inefficient. In practice, multithreading-relevant bugs are usually buried in sophisticated program flows. Meanwhile, the existing grey-box fuzzing techniques do not stress thread-interleavings which affect execution states in multithreaded programs. Therefore, mainstream grey-box fuzzers cannot effectively test problematic segments in multithreaded programs despite they might obtain high code coverage statistics. To this end, we propose MUZZ, a new grey-box fuzzing technique that hunts for bugs in multithreaded programs. MUZZ owns three novel thread-aware instrumentations, namely coverage-oriented instrumentation, thread-context instrumentation, and schedule-intervention instrumentation. During fuzzing, these instrumentations engender runtime feedback to stress execution states caused by thread interleavings. By leveraging the feedback in the dynamic seed selection and execution strategies, MUZZ preserves more valuable seeds that expose bugs in a multithreading context. We evaluate MUZZ on 12 real-world software programs. Experiments show that MUZZ outperforms AFL in both multithreading-relevant seed generation and concurrency-vulnerability detection. Further, by replaying the target programs against the generated seeds, MUZZ also reveals more concurrency-bugs (e.g., data-races, thread-leaks) than AFL. In total, MUZZ detected 8 new concurrency-vulnerabilities and 19 new concurrency-bugs. At the time of writing, 4 CVE IDs have been assigned to the reported issues.

Jiaming Ye, Mingliang Ma, Yun Lin et al.

Recent years have seen smart contracts are getting increasingly popular in building trustworthy decentralized applications. Previous research has proposed static and dynamic techniques to detect vulnerabilities in smart contracts. These tools check vulnerable contracts against several predefined rules. However, the emerging new vulnerable types and programming skills to prevent possible vulnerabilities emerging lead to a large number of false positive and false negative reports of tools. To address this, we propose Vulpedia, which mines expressive vulnerability signatures from contracts. Vulpedia is based on the relaxed assumption that the owner of contract is not malicious. Specifically, we extract structural program features from vulnerable and benign contracts as vulnerability signatures, and construct a systematic detection method based on detection rules composed of vulnerability signatures. Compared with the rules defined by state-of-the-arts, our approach can extract more expressive rules to achieve better completeness (i.e., detection recall) and soundness (i.e., precision). We further evaluate Vulpedia with four baselines (i.e., Slither, Securify, SmartCheck and Oyente) on the testing dataset consisting of 17,770 contracts. The experiment results show that Vulpedia achieves best performance of precision on 4 types of vulnerabilities and leading recall on 3 types of vulnerabilities meanwhile exhibiting the great efficiency performance.

Sen Chen, Lingling Fan, Guozhu Meng et al.

Mobile banking apps, belonging to the most security-critical app category, render massive and dynamic transactions susceptible to security risks. Given huge potential financial loss caused by vulnerabilities, existing research lacks a comprehensive empirical study on the security risks of global banking apps to provide useful insights and improve the security of banking apps. Since data-related weaknesses in banking apps are critical and may directly cause serious financial loss, this paper first revisits the state-of-the-art available tools and finds that they have limited capability in identifying data-related security weaknesses of banking apps. To complement the capability of existing tools in data-related weakness detection, we propose a three-phase automated security risk assessment system, named AUSERA, which leverages static program analysis techniques and sensitive keyword identification. By leveraging AUSERA, we collect 2,157 weaknesses in 693 real-world banking apps across 83 countries, which we use as a basis to conduct a comprehensive empirical study from different aspects, such as global distribution and weakness evolution during version updates. We find that apps owned by subsidiary banks are always less secure than or equivalent to those owned by parent banks. In addition, we also track the patching of weaknesses and receive much positive feedback from banking entities so as to improve the security of banking apps in practice. To date, we highlight that 21 banks have confirmed the weaknesses we reported. We also exchange insights with 7 banks, such as HSBC in UK and OCBC in Singapore, via in-person or online meetings to help them improve their apps. We hope that the insights developed in this paper will inform the communities about the gaps among multiple stakeholders, including banks, academic researchers, and third-party security companies.

Guozhu Meng, Yinxing Xue, Jing Kai Siow et al.

Data driven research on Android has gained a great momentum these years. The abundance of data facilitates knowledge learning, however, also increases the difficulty of data preprocessing. Therefore, it is non-trivial to prepare a demanding and accurate set of data for research. In this work, we put forward AndroVault, a framework for the Android research composing of data collection, knowledge representation and knowledge extraction. It has started with a long-running web crawler for data collection (both apps and description) since 2013, which guarantees the timeliness of data; With static analysis and dynamic analysis of the collected data, we compute a variety of attributes to characterize Android apps. After that, we employ a knowledge graph to connect all these apps by computing their correlation in terms of attributes; Last, we leverage multiple technologies such as logical inference, machine learning, and correlation analysis to extract facts (more accurate and demanding, either high level or not, data) that are beneficial for a specific research problem. With the produced data of high quality, we have successfully conducted many research works including malware detection, code generation, and Android testing. We would like to release our data to the research community in an authenticated manner, and encourage them to conduct productive research.