Against the backdrop of changing international situations and the growing importance of independently cultivating basic science talents, this paper explores the cultivation of top−tier science and technology talents in China. It proposes countermeasures including establishing an innovative training mechanism for gifted individuals, introducing world−class scholars to strengthen basic teaching and broaden horizons, providing sufficient space and time for free academic exploration, establishing evaluation criteria that emphasize original value, and building international exchange platforms. The paper points out that the next decade is a critical period for China to transform from a "major mathematical power" to a "world−leading mathematical power." It calls for collaborative efforts from multiple sectors to achieve the cultivation of indigenous top−tier science and technology talents and breakthroughs in basic sciences, thereby laying a solid foundation for China to become a world−class science and technology power.

Against the backdrop of profound restructuring of the global sci−tech industrial landscape, this study addresses the lack of systematic review of China's sci−tech industrial innovation development and the insufficient holistic analytical framework for its key breakthroughs and leading advantages. Taking the development course of China's sci−tech industry since the reform and opening−up as the main line, combined with Lu Yongxiang's theoretical discussions on scientific and technological innovation and the development of new quality productive forces, this paper systematically sorts out major innovation achievements in fields such as energy and power, quantum information, artificial intelligence, aerospace engineering, transportation infrastructure, advanced manufacturing, and national defense technology over the past two decades. It analyzes key driving factors including policy support, talent development, and industrial transformation, and reveals China's innovation development path and evolution law from following imitation to running neck and neck with global leaders and taking the lead. The study shows that China has achieved technological breakthroughs and taken the lead in industries such as renewable energy, high−speed rail networks, aerospace engineering, quantum computing, and the digital economy, forming a full−chain innovation model covering basic research, technological innovation, and large−scale industrialization. The large−scale development of green energy, high−end equipment, and new materials industries has become a core driving force for the development of new quality productive forces. The review and analysis in this study can provide reference and support for the formulation of China's sci−tech industrial development strategies, optimization of the innovation ecosystem, and layout of future cutting−edge technologies.

This review summarizes significant breakthroughs achieved over the past five years in key optical domains prioritized by the 15th Five−Year Plan, ranging from laser−driven inertial confinement fusion and intelligent photonics to quantum optics and high−precision optical manufacturing. In laser−driven inertial confinement fusion, new records in fusion energy production continue to be set, and inertial fusion energy research is advancing toward engineering application. In optics and artificial intelligence, photonic chips show strong potential in computing performance and energy efficiency, while artificial intelligence is becoming increasingly integrated with computational imaging and optical design. In quantum optics, high−performance quantum light sources are being further developed, and quantum networks and quantum metrology are progressing toward practical deployment. In high−precision optical manufacturing, research efforts are centered on extreme ultraviolet lithography and laser precision manufacturing. Optics is rapidly evolving from a stage centered on fundamental principles and device innovation to one emphasizing system integration and engineering applications. This transition is expected to exert profound influence on energy, manufacturing, and precision measurement, establishing optics as a foundational technology that supports future societal development and industrial transformation.

Artificial intelligence (AI) is pressing the accelerator button to break through the bottleneck of biomedicine research and development, profoundly changing the medical model. This paper provides an overview of the competitive landscape of AI−enabled biomedicine globally. European and American countries are accelerating the application of AI in the biomedicine and have established a significant first−mover advantage. Then, this paper analyzes the current status of AI−enabled biomedicine in China, which is in the early stages of rapid development, while facing severe challenges such as the heavy burden of major diseases and decoupling from international chains. Moreover, the main issues identified include the lack of high−quality biomedicine data, insufficient underlying original innovation in AI, imperfect policy regulation and industrial chain ecosystem construction, and a shortage of composite talents. Finally, we suggest promoting the construction of a national research−oriented biomedicine data platform, increasing the original innovation of AI for biomedicine, promoting the landing and application of AI in the fields of traditional Chinese medicine, biomedicine, and brain computer interfaces, and increasing the cultivation of versatile talents.

As design costs at advanced process nodes grow exponentially, the chiplet architecture is regarded as a key path to break through the economic bottleneck of monolithic integration. However, the current chiplet industry faces a threefold dilemma of "suspended standards, high barriers, and a sparse ecosystem": although interconnect standards such as the Universal Chiplet Interconnect Express (UCIe) consortium are evolving rapidly, a vast engineering gap remains between the standard specification and a manufacturable physical implementation; leading vendors generally rely on in−house proprietary solutions, so the standards lack tape−out validation from real products and are difficult to refine; and the high development cost deters small and medium−sized enterprises, leaving ecosystem participants extremely scarce. This paper proposes the concept of a "chiplet library", a hybrid rapid−verification system based on an FPGA prototyping platform, a chiplet−interconnect physical−layer (PHY) verification card, and EDA simulation compensation. The field−programmable gate array (FPGA) platform carries the programmable verification of chiplet functional logic and the digital layer of the interconnect protocol; the physical−layer verification card uses a real advanced−packaging process to provide direct measurement of in−package interconnect channels; and electronic design automation (EDA) simulation tools compensate for the complete on−chip (OC) physical−field environment missing on the verification card—including power distribution network (PDN) noise, thermal coupling, and cross−die interference—while supporting extrapolation to packaging−process variants. Together the three form a complete verification chain. The known limitations of this approach, in terms of operating−frequency differences, interface intellectual property (IP) partitioning, and the lack of a complete system−on−chip (SoC) physical environment, are analyzed and corresponding strategies are proposed. The industrial significance of the chiplet library is then discussed with respect to driving the iterative practice of standards, revitalizing existing chip resources, and accelerating the maturation of incremental chiplet−architecture development.

As a frontier technology in the current artificial intelligence era, generative artificial intelligence (GAI) is profoundly reshaping society across multiple domains. Concomitant with the proliferation of GAI applications are the emerging security challenges at both technical and social levels. To better understand the technical and social issues brought about by GAI, it is imperative to conduct a systematic and comprehensive investigation into existing security challenges, developed countermeasures, and future directions. This paper conducts a systematic survey spanning the entire lifecycle of GAI—namely, training security, inference security, and derived security—which correspond to model training, model inference, and model application, respectively. Formal models for GAI training and inference are developed to articulate security vulnerabilities, threat surfaces, and associated factors. An in−depth analysis covers typical attacks and corresponding countermeasures at both the training and inference stages. Derived security is also investigated, referring to security risks arising from GAI applications, including misinformation, social fairness concerns, individual privacy threats, and more, followed by a review of relevant countermeasures. Future directions in GAI security governance are discussed, emphasizing controllable and trustworthy generation mechanisms, security evaluation benchmarks, accountability mechanisms, and regulatory compliance frameworks. Overall, this article conducts a comprehensive investigation into the security risks and defense measures of GAI and constructs a security research framework that covers the entire lifecycle of GAI, as well as its research status at both technical and social levels.

Aerogels, characterized by their ultralow density and high porosity, exhibit exceptional physical and chemical properties, and have demonstrated broad application potential in thermal protection, catalysis, energy storage, sensing, and composite materials. Owing to these advantages, aerogels were once selected by the International Union of Pure and Applied Chemistry as one of the "Top Ten Emerging Technologies in Chemistry". Conventional aerogels are predominantly fabricated via ice−templating and sol–gel routes. However, with the growing demand for large−scale and engineering−oriented applications, the low fabrication efficiency and high energy consumption have become critical bottlenecks. In recent years, two−dimensional confined chemistry foaming has emerged as a promising strategy for the rapid and controllable fabrication of aerogel materials. The intrinsically resulting dome−celled microstructures not only enable efficient processing but also significantly enhance mechanical performance, thereby opening new avenues for expanding the structural and functional design space of traditional aerogels. This paper systematically reviews recent progress in 2D confined foaming of dome−celled aerogels, with a particular focus on fabrication strategies, underlying formation mechanisms, multiscale structural mechanics, and multifunctional applications. Furthermore, the critical challenges facing this class of aerogels in structure−property regulation and industrial production are analyzed, and their application prospects in acoustic regulation, intelligent sensing, and extreme thermal protection are highlighted. This review aims to provide a fundamental theoretical basis for improving aerogel performance and promoting their engineering deployment, thereby accelerating their utilization in extreme environments such as aerospace and specialized industrial applications.

Driven by the global energy transition and the "dual carbon" goals, intelligent nuclear power has become the core path for the nuclear energy industry to break through the bottlenecks in safety and economy. Nuclear power safety constitutes the fundamental prerequisite for the development of intelligent nuclear power, while intelligent control stands as its most crucial feature. This paper proposes constructing an open and collaborative digital−intelligent application ecosystem for nuclear power, with the digital−intelligent platform as the foundation and digital−intelligent technologies as the core. It aims to gradually empower and integrate with nuclear power control systems, thereby establishing a trinity intelligent nuclear power system characterized by "safety as the foundation, ecology as the wing, and efficiency as the core". Meanwhile, it puts forward a "three−step" plan for the development of intelligent nuclear power over the next decade: establishing an improved intelligent nuclear power standard system within 3 years, launching engineering demonstration applications within 5 years, and completing the construction of a non−safety class intelligent nuclear power ecosystem within 10 years. The paper presents a top−level physical architecture of intelligent nuclear power rooted in nuclear power operation safety, and develops a digital−intelligent platform with modular, process−oriented, and low−code configurable capabilities. By adopting digital−intelligent technologies such as cloud computing, big data, the Internet of Things, mobile Internet, and artificial intelligence, it realizes intelligent services including advanced perception, one−click start−stop, and autonomous decision−making. Ultimately, an innovation−driven model of "platform+business ecosystem+mechanistic model+data−driven+artificial intelligence" is formed, providing directional guidance for energy security and the digital−intellectual transformation of the nuclear energy industry.

Coal−uranium co−mining is a strategic measure to resolve the structural contradiction between "orderly coal phase−down and large−scale nuclear energy development" during China's energy transition. This paper systematically analyzes the strategic connotation, key technology system, and development path of coal−uranium co−mining. The study first elucidates the strategic value of coal−uranium co−mining from three dimensions—resource security, regional transformation, and resource utilization efficiency—demonstrating its necessity as a key pathway for a smooth transition of the energy system. Subsequently, a framework of key technology systems is constructed, centered on four core modules: precise exploration and collaborative evaluation, collaborative design and optimization, safety and environmental disturbance control, and intelligent empowerment. The paper focuses on the differentiated development strategies and core technical challenges for two typical symbiotic structures: "coal−over−uranium" and "uranium−over−coal". On this basis, a three−phase development path is proposed: "demonstration and exploration–large−scale promotion–deep integration". Corresponding safeguard measures are put forward from three aspects: top−level design, technological innovation, and policies and standards. The research indicates that coal−uranium co−mining is a strategic choice for achieving high−quality development of energy resources and ensuring national uranium resource security, requiring national−level overall planning and systematic promotion.

Space science embodies the dual mission of exploring the frontiers of knowledge and generating broad societal value, making it an important strategic source for seizing the commanding heights of aerospace science and technology, fostering new quality productive forces, and supporting Chinese modernization. Based on a review of international practices and China's recent progress in space science, this paper analyzes three mechanisms through which space science generates new growth drivers: extreme−goal traction, technology spillover, and cognitive primacy. It argues that China still faces three major constraints: insufficient supply of major space science missions, shortcomings in payload capability, and an underdeveloped innovation−to−industry conversion ecosystem. Looking ahead to the 15th Five−Year Plan period, the paper proposes three strategic pathways: strengthening the layout of major missions, enhancing key payload and foundational capabilities, and optimizing the conversion ecosystem, so as to better translate space science from frontier exploration into technological, industrial, and strategic advantages.

As a new type of production factor, data elements have become an important driving force for promoting high−quality development of agriculture. Based on provincial panel data from 2011 to 2024, this study constructs an evaluation system for high−quality agricultural development from the two dimensions of efficiency and structure. Using a two−way fixed effects model, it empirically examines the causal effect of data elements on high−quality agricultural development and explores the underlying mechanisms through which data elements promote such development. The findings reveal that data elements significantly improve both the efficiency and structure of high−quality agricultural development. The efficiency of agricultural labor allocation and the upgrading of industrial structure serve as positive mediators, jointly facilitating the "efficiency enhancement and structural upgrading" of agriculture. The enabling effect of data elements on high−quality agricultural development is heterogeneously influenced by agricultural resource allocation efficiency, agricultural R&D and innovation capacity, and the industrial co−agglomeration index. The eastern and centra regions exhibit higher levels in terms of agricultural resource allocation efficiency, agricultural R&D and innovation capacity, and the industrial co−agglomeration index. In the western region, the effects on agricultural resource allocation efficiency and the industrial co−agglomeration index are not significant, whereas the effect on agricultural R&D and innovation capacity is significant. Based on these results, this paper proposes accelerating the deep integration of data elements with the entire agricultural industry chain, enhancing the level of digital technology empowerment of agricultural labor, and building a data−driven modern agricultural industrial system.

Preoperative accurate focal localization is a critical determinant of the precision and safety of high−intensity focused ultrasound (HIFU) therapy. Acoustic Radiation Force Impulse (ARFI) imaging and Harmonic Motion Imaging (HMI), which combine therapeutic transducers with acoustic radiation force elastography, have both been proven to improve the accuracy of HIFU therapy, but there remains controversy regarding their performance differences. To address this issue, this study systematically compared the advantages and disadvantages of these two methods through in vitro and in vivo experiments. Porcine longissimus dorsi, bovine liver (in vitro), and live rabbits (in vivo) were used as experimental subjects. ARFI and HMI methods were applied for HIFU focal localization under different parameters, and the localization errors from multiple experiments were quantitatively analyzed. Results across all experimental subjects showed that the average focal error of the traditional geometric localization method was approximately 3−4 mm, while the localization error of ARFI could be stably controlled at around 1 mm, and the error of HMI was close to that of ARFI. However, ARFI technology requires a higher output power, approximately 2−3 times that of HMI. Therefore, ARFI and HMI based on acoustic radiation force elastography can both be used for preoperative HIFU focal localization. Both methods are significantly superior to the traditional geometric localization method in terms of precision, with HMI offering better safety. This study provides important and comprehensive experimental evidence for the clinical application of dynamic acoustic radiation force elastography in HIFU focal localization.

Mega science projects, with large−scale research facilities as their core carriers, span the full lifecycle from planning and construction to long−term operation and open access; their performance ultimately depends on sustained governance and robust operational mechanisms. This paper takes knowledge discovery mechanisms and organizational governance models as the main analytical lenses and conducts a comparative analysis between China and America. The results show that the United States, relying on a distributed collaboration network connecting national laboratories, universities, and industry, achieves iterative cycles of open access and technology transfer through mature user programs, clearer IP rules, and contract−based management. China, leveraging a new whole−of−nation system and clustered national science centers, demonstrates strengths in strategic alignment and system integration, and promotes technological spillovers and industry coupling through mechanisms such as "spin−offs along the way." Looking ahead to the 15th Five−Year Plan period, mega science projects are expected to exhibit intertwined trends including more complex strategic objectives, ecosystem−based governance, refined operational management, AI and data−driven discovery, and increasingly multi−layered international collaboration. Accordingly, this paper proposes building an integrated "strategy–platform–ecosystem" framework for construction and governance, strengthening operation−phase funding, user governance, and performance evaluation, institutionalizing data governance and AI capabilities, prudently advancing diversified financing and cost−sharing, and enhancing international cooperation and participation in rule−making.

Lu Qikeng (1927—2015) was a principal pioneer of the discipline of several complex variables in China, and one of the core inheritors of the Hua Luogeng School. Despite physical disabilities, he achieved self−taught academic success. Together with Hua Luogeng, he co−established the disciplinary system of several complex variables in China, fostered China's first research team in this field and more than 30 academic leaders. His academic achievements were remarkable: he proposed the "Lu Qikeng Conjecture" and proved the "Lu Qikeng Theorem", with related research leading the West by more than a decade. He took the lead in revealing the connection between gauge fields and fiber bundle connection theory. As a forerunner of academic exchanges, he promoted in − depth mathematical exchanges between China and foreign countries. He was honored with many awards, including the first Hua Luogeng Mathematics Prize. The research achievements in several complex variables, to which he contributed to the foundation and development, were listed by General Secretary Xi Jinping as important scientific achievements since the founding of the People's Republic of China. Throughout his life, he practiced the concepts of serving the country through science and devoting himself to educating people, composing an inspiring life movement of perseverance.