Xiangyu Liu

Android’s vast ecosystem and expansive API surfaces pose a serious challenge to static application security testing (SAST) tools. Mainstream tools such as MobSF, APKHunt, and AUSERA mainly rely on manually crafted rules. Crafting these rules demands considerable effort, yet they still cannot cover every security issue. When Android introduces new APIs, changes its permission model, or revises other security policies, the rules soon fall behind. Without constant maintenance, false positives grow, and true vulnerabilities slip through. Recently released LLM-based detectors are easy to use and potentially support a wide range of vulnerability types, but their findings often lack clear explanations and suffer from high false-positive rates. In this paper, we present RULEDROID, a new framework that leverages LLMs to automatically generate Semgrep-compatible static detection rules from up-to-date official Android security documentation. RULEDROID tackles the limits of pure LLM detection by combining (i) Retrieval-Augmented Generation (RAG), which grounds model outputs in trusted documents, and (ii) a modular workflow that decomposes rule synthesis into welldefined stages. The resulting rules are then applied with proven static-analysis techniques, ensuring consistent and explainable results. We evaluated RULEDROID on three public benchmark datasets. Based on its large and precise rule set, RULEDROID achieved higher coverage and accuracy than traditional SAST tools, and sharply reduced false positives compared with direct LLM scanning. When applied to real-world apps, RULEDROID discovered multiple new vulnerabilities, resulting in 57 CVE IDs being assigned. These results show that RULEDROID combines the broad vulnerability coverage of LLMs with the precision of static analysis, delivering a fully automated docs-to-rules solution for Android security testing.

Domain generalization (DG) aims to learn from multiple source domains a model that can generalize well on unseen target domains. Existing DG methods mainly learn the representations with invariant marginal distribution of the input features, however, the invariance of the conditional distribution of the labels given the input features is more essential for unknown domain prediction. Meanwhile, the existing of unobserved confounders which affect the input features and labels simultaneously cause spurious correlation and hinder the learning of the invariant relationship contained in the conditional distribution. Interestingly, with a causal view on the data generating process, we find that the input features of one domain are valid instrumental variables for other domains. Inspired by this finding, we propose an instrumental variable-driven DG method (IV-DG) by removing the bias of the unobserved confounders with two-stage learning. In the first stage, it learns the conditional distribution of the input features of one domain given input features of another domain. In the second stage, it estimates the relationship by predicting labels with the learned conditional distribution. Theoretical analyses and simulation experiments show that it accurately captures the invariant relationship. Extensive experiments on real-world datasets demonstrate that IV-DG method yields state-of-the-art results.

The blockchain-powered decentralized applications and systems have been widely deployed in recent years. The decentralization feature promises users anonymity, security, and non-censorship, which is especially welcomed in the areas of decentralized finance and digital assets. From the perspective of most common users, a decentralized ecosystem means every service follows the principle of decentralization. However, we find that the services in a decentralized ecosystem still may contain centralized components or scenarios, like third-party SDKs and privileged operations, which violate the promise of decentralization and may cause a series of centralized security risks. In this work, we systematically study the centralized security risks existing in decentralized ecosystems. Specifically, we identify seven centralized security risks in the deployment of two typical decentralized services – crypto wallets and DApps, such as anonymity loss and overpowered owner. Also, to measure these risks in the wild, we designed an automated detection tool called Naga and carried out large-scale experiments. Based on the measurement of 28 Ethereum crypto wallets (Android version) and 110,506 on-chain smart contracts, the result shows that the centralized security risks are widespread. Up to 96.4% of wallets and 83.5% of contracts exist at least one security risk, including 260 well-known tokens with a total market cap of over $98 billion.

Boxin Wang, Chejian Xu, Xiangyu Liu, Yu Cheng, Bo Li. Findings of the Association for Computational Linguistics: NAACL 2022. 2022.

Despite that deep neural networks (DNNs) have achieved enormous success in many domains like natural language processing (NLP), they have also been proven to be vulnerable to maliciously generated adversarial examples. Such inherent vulnerability has threatened various real-world deployed DNNs-based applications. To strength the model robustness, several countermeasures have been proposed in the English NLP domain and obtained satisfactory performance. However, due to the unique language properties of Chinese, it is not trivial to extend existing defenses to the Chinese domain. Therefore, we propose AdvGraph, a novel defense which enhances the robustness of Chinese-based NLP models by incorporating adversarial knowledge into the semantic representation of the input. Extensive experiments on two real-world tasks show that AdvGraph exhibits better performance compared with previous work: (i) effective – it significantly strengthens the model robustness even under the adaptive attacks setting without negative impact on model performance over legitimate input; (ii) generic – its key component, i.e., the representation of connotative adversarial knowledge is task-agnostic, which can be reused in any Chinese-based NLP models without retraining; and (iii) efficient – it is a light-weight defense with sub-linear computational complexity, which can guarantee the efficiency required in practical scenarios.