Group 1 - The core issue in the domestic securities industry is the increasing complexity and difficulty in detecting abnormal trading behaviors due to the significant growth in trading products and strategies [1][2] - Traditional monitoring systems are lagging in technical response, primarily relying on second-level data, which fails to capture millisecond-level trading behaviors, leading to monitoring blind spots [2][3] - There are dual risks associated with core risk control modules that depend on foreign technology architectures, which do not meet local regulatory and data security requirements [2][3] Group 2 - The current state of risk control in the securities industry is hindered by a lack of comprehensive domestic hardware and software integration, limiting the effectiveness of risk management systems [3] - Existing systems often rely on traditional machine learning models, lacking the ability to comprehensively identify trading behavior patterns and complex relationships [3] Group 3 - A proposed multi-technology integration framework includes FPGA, distributed technology, large models, and knowledge graphs, which can create a closed-loop risk control mechanism from data collection to prevention [4][5] - FPGA technology allows for low-latency data collection, enabling real-time capture of market microstructure changes and reducing warning delays to milliseconds [4] - Distributed technology addresses the explosive growth of order data during trading peaks through parallel computing, enhancing real-time data processing capabilities [6] Group 4 - The integration of large models and knowledge graphs enhances the ability to model and analyze trading behaviors, improving the accuracy of abnormal trading detection [7] - A fully domestic risk control system is proposed, covering all critical components from hardware to algorithms, to ensure security and reduce external dependencies [8] Group 5 - Challenges in implementing multi-technology integration include difficulties in adapting different technological systems and the lack of standardized solutions for domestic FPGA chips [9][10] - There is a significant shortage of composite talents who possess knowledge in finance, information technology, AI algorithms, and hardware engineering, which hampers the development and operation of risk control systems [10] Group 6 - Recommendations for enhancing the application of advanced technologies in abnormal trading risk control include establishing a collaborative innovation platform, improving domestic industrial chain support policies, and fostering composite talent development [11][12][14] - Pilot demonstrations and ecosystem cultivation are suggested to validate the application of integrated risk control systems in securities firms, promoting technology sharing and collaboration [15]