The complexity of code reviews has driven efforts to automate review comments, but prior approaches oversimplify this task by treating it as snippet-level code-to-text generation and relying on text similarity metrics like BLEU for evaluation. These methods overlook repository context, real-world merge request evaluation, and defect detection, limiting their practicality. To address these issues, we explore the full automation pipeline within the online recommendation service of a company with nearly 400 million daily active users, analyzing industry-grade C++ codebases comprising hundreds of thousands of lines of code. We identify four key challenges: 1) capturing relevant context, 2) improving key bug inclusion (KBI), 3) reducing false alarm rates (FAR), and 4) integrating human workflows. To tackle these, we propose 1) code slicing algorithms for context extraction, 2) a multi-role LLM framework for KBI, 3) a filtering mechanism for FAR reduction, and 4) a novel prompt design for better human interaction. Our approach, validated on real-world merge requests from historical fault reports, achieves a 2× improvement over standard LLMs and a 10× gain over previous baselines. While the presented results focus on C++, the underlying framework design leverages language-agnostic principles (e.g., AST-based analysis), suggesting potential for broader applicability.

Existing methods in RFID systems often employ presence or absence fashion to detect the tags' motions, so they cannot meet motion detection requirement in many applications. Our recent observations suggest that the signal strength backscattered from the tag is hypersensitive to its position, inspiring us to perceive the tag motion through its radio signal strength changes. Motion perception is not trivial and challenged by weak stability of strength in that any other interference or noise may incur significant changes as well, resulting in high false positives. To tackle this issue, we propose to model the strength via the Mixture of Gaussian Model (MoG). The problem is thus converted to foreground segment in computer vision with the help of Strength Image, where the technique of MoG based background subtraction is employed. We then implement a prototype using commercial off-the-shelf products. The evaluation results show that the slightest tag motion (~ 10cm) can be precisely perceived, and the accuracy is up to 92.34% while the false positive is suppressed under 0.5%.