Automated GUI testing plays a crucial role for smartphone vendors who have to ensure that the widely used mobile apps-that are not essentially developed by the vendors-are compatible with new devices and system updates. While existing testing techniques can automatically generate event sequences to reach different GUI views, inputs such as strings and numbers remain difficult to generate, as their generation often involves semantic understanding of the app functionality. Recently, Large Language Models (LLMs) have been successfully adopted to generate string inputs that are semantically relevant to the test case. This paper evaluates the LLM-based input generation in the industrial context of vendor testing of both in-house and 3rd party mobile apps. We present DROIDFILLER, an LLM based input generation technique that builds upon existing work with more sophisticated prompt engineering and customisable context retrieval. DROIDFILLER is empirically evaluated using a total of 120 textfields collected from a total of 45 apps, including both in-house and 3rd party ones. The results show that DROIDFILLER can outperform both vanilla LLM based input generation as well as the existing resource pool approach. We integrate DROIDFILLER into the existing GUI testing framework used at Samsung, evaluate its performance, and discuss the challenges and considerations for practical adoption of LLM-based input generation in the industry.

As one of autonomous vehicles, conditional autonomous vehicles is expected to become popular in the near future. Conditional autonomous vehicles can send a take-over request (TOR) to a driver, and if they are immersed in non-driving-related tasks (NDRT), they will struggle to accommodate this request. Previous studies have shown that providing augmented reality (AR) information on traffic situations (status cues) or driver actions (command cues) can improve TOR performance. However, we are not aware of any studies comparing the types of AR cues (state versus command cues) and their interactions with environmental factors. Therefore, the current study investigated this and evaluated the TOR performance of 42 drivers. We used a 2 (environments: day and night) $\times$ 4 (AR cue types: without, status, command, and combined cues) mixed-subject experimental design, and dependent measures included driving, cognitive, and NDRT performances. The results suggest that overall driving and cognitive performance were significantly improved by the command AR cue. In contrast, the status AR cue improved the TOR performance in nighttime environments. The performance of AR cues can vary depending on environmental factors, and AR cue designs for autonomous vehicles should consider this interaction for successful collaboration between drivers and vehicles.