Recent advancements in end-to-end autonomous driving systems (ADSs) underscore their potential for perception and planning capabilities. However, challenges remain. Complex driving scenarios contain rich semantic information, yet ambiguous or noisy semantics can compromise decision reliability, while interference between multiple driving tasks may hinder optimal planning. Furthermore, prolonged inference latency slows decision-making, increasing the risk of unsafe driving behaviors. To address these challenges, we propose ExpertAD, a novel framework that enhances the performance of ADS with Mixture of Experts (MoE) architecture. We introduce a Perception Adapter (PA) to amplify task-critical features, ensuring contextually relevant scene understanding, and a Mixture of Sparse Experts (MoSE) to minimize task interference during prediction, allowing for effective and efficient planning. Our experiments show that ExpertAD reduces average collision rates by up to 20% and inference latency by 25% compared to prior methods. We further evaluate its multi-skill planning capabilities in rare scenarios (e.g., accidents, yielding to emergency vehicles) and demonstrate strong generalization to unseen urban environments. Additionally, we present a case study that illustrates its decision-making process in complex driving scenarios.

Open source software (OSS) has become the foundation of modern applications, but its transitive dependencies make it especially vulnerable to supply chain attacks. One common tactic is to inject malicious code into third-party packages. NPM, in particular, due to its widespread use and large volume of packages, has become the popular target of malicious code injection. While various detectors have been proposed, they suffer three limitations, i.e., inadequate behavior modeling of obfuscated code, ignoring object-centric features of JavaScript, and lack of synergy between static and dynamic analysis. These limitations lead to imprecise modeling of program behavior and hinder detection effectiveness.To address these limitations, we propose ProfMal to identify malicious NPM packages, which leverages the synergy between static and dynamic analysis to construct behavior graphs for each package. Specifically, our static analysis constructs the behavior graphs through object-sensitive analysis, while identifying sensitive API calls and locating statically unresolved calls. Our dynamic analysis augments the behavior graphs by resolving those statically unresolved calls. Based on these comprehensive behavior graphs, we train a graph-based classifier to identify maliciousness. Our evaluation has indicated that ProfMal achieves the highest F1-score of 92.4%, outperforming the state-of-the-arts by 6.2% to 48.8%. During a three-month real-world detection, ProfMal has detected 496 previously unknown malicious NPM packages, and all of them have been confirmed and removed from NPM.

End-to-end autonomous driving systems (ADSs), with their strong capabilities in environmental perception and generalizable driving decisions, are attracting growing attention from both academia and industry. However, once deployed on public roads, ADSs are inevitably exposed to diverse driving hazards that may compromise safety and degrade system performance. This raises a strong demand for resilience of ADSs, particularly the capability to continuously monitor driving hazards and adaptively respond to potential safety violations, which is crucial for maintaining robust driving behaviors in complex driving scenarios.To bridge this gap, we propose a resilience-oriented runtime framework, named Argus, to mitigate the driving hazards, thus preventing potential safety violations and improving the driving performance of an ADS. Argus continuously monitors the trajectories generated by the ADS for potential hazards and, whenever the EGO vehicle is deemed unsafe, seamlessly takes control via a hazard mitigator. We integrate Argus with three state-of-the-art end-to-end ADSs, i.e., TCP, UniAD and VAD. Our evaluation has demonstrated that Argus effectively and efficiently enhances the resilience of ADSs, improving the driving score of ADSs by 150.30% on average, and preventing 64.38% of the violations, with little additional time overhead.

Simulation testing has been widely adopted by leading companies to ensure the safety of autonomous driving systems (ADSs). Anumber of scenario-based testing approaches have been developed to generate diverse driving scenarios for simulation testing, and demonstrated to be capable of finding safety violations. However, there is no automated way to diagnose whether these violations are caused by the ADS under test and which category these violations belong to. As a result, great effort is required to manually diagnose violations. To bridge this gap, we propose DiaVio to automatically diagnose safety violations in simulation testing by leveraging large language models (LLMs). It is built on top of a new domain specific language (DSL) of crash to align real-world accident reports described in natural language and violation scenarios in simulation testing. DiaVio fine-tunes a base LLM with real-world accident reports to learn diagnosis capability, and uses the fine-tuned LLM to diagnose violation scenarios in simulation testing. Our evaluation has demonstrated the effectiveness and efficiency of DiaVio in violation diagnosis.