This Appendix is divided into three sections. First, Section A empirically validates the feasibility of Assumption 1. Next, in Section B, complete proofs of all the lemmas and theorems are presented. Finally, in Section C, we provide detailed settings of baselines and additional experimental results, including the hyper-parameter analysis and the percentage of rectified data. The users' true preference distribution η fulfills the relaxed Multiclass Tsybakov Condition [1] with constants C > 0, λ > 0, and α0 (0,1], such that for all α (0,α0], The feasibility of Assumption 1 relies on large C and small λ. In order to estimate the values of C and λ, the initial step is to approximate the true preference distribution η.

Sequential recommender systems (SRSs) are typically trained to predict the next item as the target given its preceding (and succeeding) items as the input. Such a paradigm assumes that every input-target pair is reliable for training. However, users can be induced to click on items that are inconsistent with their true preferences, resulting in unreliable instances, i.e., mismatched input-target pairs. Current studies on mitigating this issue suffer from two limitations: (i) they discriminate instance reliability according to models trained with unreliable data, yet without theoretical guarantees that such a seemingly contradictory solution can be effective; and (ii) most methods can only tackle either unreliable input or targets but fail to handle both simultaneously. To fill the gap, we theoretically unveil the relationship between SRS predictions and instance reliability, whereby two error-bounded strategies are proposed to rectify unreliable targets and input, respectively. On this basis, we devise a model-agnostic Bidirectional Data Rectification (BirDRec) framework, which can be flexibly implemented with most existing SRSs for robust training against unreliable data. Additionally, a rectification sampling strategy is devised and a self-ensemble mechanism is adopted to reduce the (time and space) complexity of BirDRec. Extensive experiments on four real-world datasets verify the generality, effectiveness, and efficiency of our proposed BirDRec.

As a safety-critical cyber-physical system, cybersecurity and related safety issues for Autonomous Vehicles (AVs) have been important research topics for a while. Among all the modules on AVs, perception is one of the most accessible attack surfaces, as drivers and AVs have no control over the outside environment. Most current work targeting perception security for AVs focuses on perception correctness. In this work, we propose an impact analysis based on inference time attacks for autonomous vehicles. We demonstrate in a simulation system that such inference time attacks can also threaten the safety of both the ego vehicle and other traffic participants.

--Track testing plays a critical role in the safety evaluation of autonomous driving systems (ADS), as it provides a real-world interaction environment. However, the inflexibility in motion control of object targets and the absence of intelligent interactive testing methods often result in pre-fixed and limited testing scenarios. T o address these limitations, we propose a novel 5G cloud-controlled track testing framework, Real-world Troublemaker . This framework overcomes the rigidity of traditional pre-programmed control by leveraging 5G cloud-controlled object targets integrated with the Internet of Things (IoT) and vehicle teleoperation technologies. Unlike conventional testing methods that rely on pre-set conditions, we propose a dynamic game strategy based on a quadratic risk interaction utility function, facilitating intelligent interactions with the vehicle under test (VUT) and creating a more realistic and dynamic interaction environment. The proposed framework has been successfully implemented at the T ongji University Intelligent Connected V ehicle Evaluation Base. Field test results demonstrate that Troublemaker can perform dynamic interactive testing of ADS accurately and effectively. Compared to traditional methods, Troublemaker improves scenario reproduction accuracy by 65.2%, increases the diversity of interaction strategies by approximately 9.2 times, and enhances exposure frequency of safety-critical scenarios by 3.5 times in unprotected left-turn scenarios. Index T erms --Automated driving systems, track testing, 5G, cloud-controlled object targets, interaction scenarios. HE safety of automated driving systems (ADS) must be ensured prior to their practical implementation, which requires a well-established testing framework [1]. Existing test standards, such as ISO 26262 [2], UN R157 [3], and UN R171 [4], are not sufficient to comprehensively evaluate ADS. According to the driving automation levels defined by SAE J3016 from SAE International, a high-level ADS (i.e., Level 3 or higher) is expected to perform driving tasks independently and autonomously, with the driver no longer retaining continuous control over vehicle movement [5]. While ADS has already been deployed in various countries and regions, numerous ADS traffic incidents highlight that safety testing for high-level ADS remains a critical technical challenge. In comparison to traditional vehicles and advanced driver assistance systems (ADAS), high-level ADS testing faces significant transformations and challenges, particularly in terms of both test subjects and requirements.

The simulation-based testing is essential for safely implementing autonomous vehicles (AVs) on roads, necessitating simulated traffic environments that dynamically interact with the Vehicle Under Test (VUT). This study introduces a VUT-Centered environmental Dynamics Inference (VCDI) model for realistic, interactive, and diverse background traffic simulation. VCDI is built on a Transformer-based trajectory inference model to generate trajectories for background objects. Serving the purpose of AV testing, VCDI additionally considers VUT-centered interactivity and scenario diversity using a conditional inference framework. First, the VUT future motion is taken as an augmented model input to bridge the interaction between VUT and background objects. Second, to enrich the scenario diversity, a Bayesian-network-based cost function module is designed. The module, learned in a distributional form, captures the uncertainty of the VUT's strategy, triggering various scenario evolution. Experimental results validate VCDI's trajectory-level simulation precision which outperforms the state-of-the-art trajectory prediction work. The flexibility of the distributional cost function allows VCDI to provide diverse-yet-realistic scenarios for AV testing. We demonstrate such capability by modifying the anticipation to VUT's cost-based strategy and thus achieve multiple testing scenarios with explainable background traffic evolution.

This Simulation Assessment Guidelines document is a public guidelines document developed by the Centre of Excellence for Testing & Research of AVs - NTU (CETRAN) in collaboration with the Land Transport Authority (LTA) of Singapore. It is primarily intended to help the developers of Autonomous Vehicles (AVs) in Singapore to prepare their software simulations and provide recommendations that can ensure their readiness for independent assessment of their virtual simulation results according to the Milestone-testing framework adopted by the assessor and the local authority in Singapore, namely, CETRAN and LTA respectively.

Conventional vehicles are certified through classical approaches, where different physical certification tests are set up on test tracks to assess required safety levels. These approaches are well suited for vehicles with limited complexity and limited interactions with other entities as last-second resources. However, these approaches do not allow to evaluate safety with real behaviors for critical and edge cases, nor to evaluate the ability to anticipate them in the mid or long term. This is particularly relevant for automated and autonomous driving functions that make use of advanced predictive systems to anticipate future actions and motions to be considered in the path planning layer. In this paper, we present and analyze the results of physical tests on proving grounds of several predictive systems in automated driving functions developed within the framework of the BRAVE project. Based on our experience in testing predictive automated driving functions, we identify the main limitations of current physical testing approaches when dealing with predictive systems, analyze the main challenges ahead, and provide a set of practical actions and recommendations to consider in future physical testing procedures for automated and autonomous driving functions.

User interface modeling is inherently multimodal, which involves several distinct types of data: images, structures and language. The tasks are also diverse, including object detection, language generation and grounding. In this paper, we present VUT, a Versatile UI Transformer that takes multimodal input and simultaneously accomplishes 5 distinct tasks with the same model. Our model consists of a multimodal Transformer encoder that jointly encodes UI images and structures, and performs UI object detection when the UI structures are absent in the input. Our model also consists of an auto-regressive Transformer model that encodes the language input and decodes output, for both question-answering and command grounding with respect to the UI. Our experiments show that for most of the tasks, when trained jointly for multi-tasks, VUT substantially reduces the number of models and footprints needed for performing multiple tasks, while achieving accuracy exceeding or on par with baseline models trained for each individual task.