Searcharxiv⌕ Search

arXiv subjects

Zhengyang Zhu

Publications and source records attributed to Zhengyang Zhu.

3 recordsLinked to original sources

HarnessSecurity-Bench: Do Security Mechanisms Really Protect Coding Agent Harnesses?

Coding agent harnesses mediate tool use and authorize actions, yet their security mechanisms and runtime effects remain incompletely characterized. We present HarnessSecurity, the first systematic empirical study and benchmark of open- and closed-source coding agent harnesses. First, we derive a ten-mechanism taxonomy and then assess 400 harness-mechanism cells using independent ratings by researchers and large language model (LLM) judges. We find that about half of confirmed mechanism implementations are opt-in, while closed-source harnesses exhibit substantial evidence gaps. Second, we introduce HarnessSecurity-Bench, a benchmark of 23 tasks across five attack surfaces without sacrificing legitimate task requirements. Using separate deterministic oracles to measure task utility and attack effects with security setting comparisons, we evaluate nine mechanisms across six leading harnesses: Claude Code, Codex CLI, Gemini CLI, gptme, Qwen Code, and GitHub Copilot. Under a controlled LLM baseline GLM-5.2, we conduct 2,500 trials, recording 81,155 tool calls and over 2.2 billion tokens. Enabling auto-approve increases utility and raises attack success from 29.2% to 95.6%. Network isolation and read-only mode reduce attack effects with substantial utility losses, while command allowlisting and command denylisting reduce attack effects with a small utility loss and a utility gain, respectively. Task-level cases show that restrictions on a shared capability can obstruct both legitimate and malicious operations, and that allowed tools or commands can leave unauthorized operations reachable through alternative execution paths. Harness providers should make security settings verifiable, test alternative execution paths to protected operations, and assess attack effects alongside task utility and execution costs.

cs.CR↗

Software Engineering Practice in the Virtual Reality/Augmented Reality Software Life Cycle: A Systematic Survey

Leveraging diverse virtual reality (VR) technologies, VR provides users with an immersive experience in a computer-generated 3D virtual environment. In contrast, augmented reality (AR) superimposes virtual objects (such as images, text, animations, etc.) onto users' real scenes. Both VR and AR have received increasing attention due to the maturity of computer graphics, computer vision, and artificial intelligence. VR and AR have gained rapid market growth in both hardware and software. In the early stages, VR/AR-related research mainly focused on hardware (e.g., head-mounted displays and sensor-equipped controllers). Recently, VR/AR software has become a hot topic due to the mature VR/AR ecosystem. However, there is no comprehensive work to summarize software engineering practices in VR/AR. To bridge this gap, this paper aims to provide a systematic survey summarizing the state-of-the-art research on VR/AR from the perspective of software engineering practices. Considering the general VR/AR software life cycle, we categorize the relevant research based on the VR/AR software life cycle, including requirement engineering and planning, software design, software development, software testing, software deployment, software maintenance and updates, and software evaluation and feedback. As for each stage above, we summarize relevant research on VR/AR for this stage, including the design of technical routes, the construction of datasets, and the design of evaluation metrics. We also enumerate existing empirical studies and make comparisons with conventional software (e.g., mobile/desktop application development). In summary, our article can help researchers gain a comprehensive understanding of the achievements of the existing software development techniques in the VR/AR field and also provide insights, which may facilitate the benign development of VR/AR software ecosystem in the future.

cs.SE↗

VRExplorer: A Model-based Approach for Semi-Automated Testing of Virtual Reality Scenes

With the proliferation of Virtual Reality (VR) markets, VR applications are rapidly expanding in scale and complexity, thereby driving an urgent need for assuring VR software quality. Different from traditional mobile applications and computer software, VR testing faces unique challenges due to diverse interactions with virtual objects, complex 3D virtual environments, and intricate sequences to complete tasks. All of these emerging challenges hinder existing VR testing tools from effectively and systematically testing VR applications. In this paper, we present VRExplorer, a novel model-based testing tool to effectively interact with diverse virtual objects and explore complex VR scenes. Particularly, we design the Entity, Action, and Task (EAT) framework for modeling diverse VR interactions in a generic way. Built upon the EAT framework, we then present the VRExplorer agent, which can achieve effective scene exploration by incorporating meticulously designed path-finding algorithms into Unity's NavMesh. Moreover, the VRExplorer agent can also systematically execute interaction decisions on top of the Probabilistic Finite State Machine (PFSM). Experimental evaluation on 11 representative VR projects shows that VRExplorer consistently outperforms the state-of-the-art (SOTA) approach VRGuide by achieving significantly higher coverage and better efficiency. Specifically, VRExplorer yields up to 122.8% and 52.8% improvements over VRGuide in terms of executable lines of code (ELOC) coverage and method (function) coverage, respectively. Furthermore, ablation results also verify the essential contributions of each designed module. More importantly, our VRExplorer has successfully detected two functional bugs and one non-functional bug from real-world projects.

cs.SE↗