High−quality data acts as a pivotal driver for the intelligent transformation of research paradigms in materials science. However, the structural deficiencies in the current materials data production system not only lead to quality issues such as data fragmentation, siloing, format heterogeneity and incomplete records, but also result in an excessive transfer of quality control responsibilities to downstream users due to weak producer accountability and the absence of a systematic governance framework. Although users attempt to enhance data governance effectiveness by integrating domain knowledge into various corresponding methods, their limited insight into data production mechanisms hinders the generalizability of their governance strategies and the reliability of results. To this end, this paper focuses on the sustainable generation of high−quality materials data as the central theme, systematically examining the causes of data quality degradation across the data full lifecycle stages, and outlining a comprehensive materials data quality control framework. It aims to guide parties involved in data production and use to reevaluate the entire process from data acquisition and management to application, while providing referential insights for implementation pathways that promote efficient delivery of high−quality materials data, thereby laying foundational support for cultivating a materials data trusted ecosystem.

At its core, the race in AI for Science is a race for talent. To grasp the full scope of this transformation, we must look beyond hard inputs such as data, computing power, and algorithms, and recognize that a softer dimension, innovation culture, is being reshaped at a systemic level. The emerging paradigm of human, AI cognitive collaboration is fundamentally altering how scientific knowledge is produced and how innovation unfolds. In this shift, defining the right problems matters more, and the need for cross−disciplinary, composite talent has become acute. Open−source practices have turned into the primary mode of research collaboration, while the accelerating pace of technological iteration makes long term commitment and tolerance for failure more valuable than ever. Widely accessible AI tools are empowering the public to engage in innovation, and ethical competence is now an integral part of innovative capability itself. The United States' recent Genesis Mission and its Science: A New Golden Age report both underscore the importance of investing in people. An innovation culture aligned with the AI era will unlock the full potential of AI for Science. Moving forward, cultivating this new culture must serve as a strategic pillar for achieving greater self-reliance and strength in science and technology. This requires focused efforts to reshape talent development systems, build open and collaborative ecosystems, reform evaluation and incentive mechanisms, strengthen technology ethics governance, and systematically enhance AI competency across the whole of society.

Rice green manure, as an agro−biological practice that simultaneously provides nutrient substitution and carbon sequestration/emission reduction benefits, holds strategic significance for concurrently safeguarding food security and advancing the green low−carbon transition in agriculture. Based on an "evolution–mechanism–pathway" framework, we systematically traces the paradigm shift from soil fertility enhancement and resource−use optimization toward ecological synergy in rice green manure research spanning 1928–2025. Critical knowledge gaps persist regarding synergistic multi−dimensional benefit quantification frameworks, micro−scale biogeochemical driving mechanisms, and digital integrated management technologies. Future priorities urgently call for: (1) a multi−dimensional evaluation framework for rice green manure cultivation grounded in water–carbon–nitrogen synergy; (2) the integration of multi−omics and isotope tracing approaches to elucidate key biogeochemical mechanisms; (3) the development of an integrated space–aerial–ground precision management system to overcome regional heterogeneity constraints; and (4) the refinement of measurement, reporting, and verification protocols to facilitate the inclusion of green manure carbon sinks into market−oriented ecological compensation mechanisms, thereby providing theoretical and technical support for the "emission reduction without yield reduction" strategy in agriculture.

The public−key cryptosystems represented by Rivest−Shamir−Adleman (RSA) and elliptic curve cryptography (ECC) are facing disruptive threats from scalable fault−tolerant quantum computing, and the "harvest−now−decrypt−later" attack pattern further underscores the urgency of post−quantum cryptographic (PQC) migration. In 2024, the U.S. National Institute of Standards and Technology (NIST) formally released the first batch of PQC algorithm standards, launching the quantum−safe upgrade of the global digital security infrastructure. Over the past five years, PQC has made important progress in the dynamic assessment of theoretical security boundaries, the global competition over algorithm standards, and the systematic exploration of industry migration paths. China has also accumulated staged results in the construction of the next−generation commercial cryptography (ShangMi, SM) standard system and pilot applications in key industries. This paper systematically reviews the development history, cutting−edge research breakthroughs and global standardization landscape of PQC, with a focus on the migration practice and exploration in the financial industry, and puts forward policy recommendations such as designating cryptographic agility as a mandatory architectural specification for critical information infrastructure and accelerating the coordination of SM and PQC dual−stack standards to establish a unified migration framework, providing a reference for China's next−stage technical research and standardization deployment.

Post−quantum cryptography, as a key technology against possible threats from quantum computing, is now becoming the foundation of next−generation cryptographic systems. The global migration to PQC is underway, with many countries actively developing new standards. Among the various candidates, lattice−based cryptography algorithms are popular due to their efficiency and small key sizes. Security analysis is crucial for ensuring the reliability of PQC. This paper focuses on the security analysis of lattice−based cryptography. We first review the development of lattice attacks and their theoretical background. Then, we study the main memory bottleneck in the best current attack method, called sieving, especially when used at cryptographic scales. By improving both the algorithm and its implementation, we achieved some new record results. Our tests show that the previous view of memory efficiency as a critical factor in lattice cryptography security estimation is significantly flawed. Based on this, we provide recommendations for choosing parameters in PQC migration and deployment, especially for critical infrastructure like banking.

This article studies the current work on post−quantum cryptography (PQC) migration in the financial and banking industry both domestically and internationally. It provides an overview of the PQC algorithm solicitation and standardization activities in the United States, introduces the relevant guidelines for PQC migration in the United States, and presents the framework design for PQC migration. Finally, a PQC migration framework applicable to China's banking industry was highlighted. On this basis, this paper reviews the origin of the PQC concept and the algorithm standardization process, as well as the current status of international PQC migration led by the National Cybersecurity Center of Excellence (NCCoE) of the United States. It then analyzes the NCCoE PQC migration framework built upon the Cybersecurity Framework (CSF) 2.0, which covers five key building blocks: quantum−safe governance, data asset inventory, management tools, risk management, and crypto−agile migration. Based on the National Key R&D Program project "Research on PQC Migration Technology for the Banking Industry and Its Critical Infrastructure Information Systems" (Project 055), a Data−drivEn Pqc−Agility migraTion framework in banking indUstRiEs (DEPATURE) is proposed, which forms a closed loop consisting of five interlocking phases: cryptographic module discovery, cryptographic asset inventory compilation, quantum−safe risk assessment, PQC agile migration, and security effectiveness verification, and finally outputs the quantum−safe governance status. The development trend of PQC migration in the international financial and banking industry is also discussed.

Post−quantum cryptography (PQC) is emerging as a critical technology for protecting financial systems against the threats posed by quantum computing and is expected to gradually replace conventional public−key cryptography. This paper systematically examines the systemic quantum threats facing the financial sector, the evolution of global PQC regulatory frameworks and standardization efforts, and the critical role of crypto−agility in facilitating a secure and timely transition. With major economies targeting the 2030–2035 period for phasing out vulnerable public−key cryptographic schemes, crypto−agility is becoming essential for financial institutions to meet emerging regulatory and compliance requirements. We further examine PQC migration across critical financial infrastructures, including central−bank payment systems, the SWIFT network, and major commercial banks, with particular attention to key technical challenges such as message−size expansion, computational overhead under hybrid cryptographic schemes, and out−of−band certificate distribution. The BIS Project Leap experiments demonstrate the functional feasibility of integrating PQC into central−bank payment infrastructures. However, the substantial message expansion caused by post−quantum signatures such as ML−DSA, along with the additional verification overhead of hybrid schemes, poses significant constraints for high−throughput clearing and settlement systems. SWIFT, meanwhile, has laid out a migration plan with SwiftNet 8.0 as its key transition node. Finally, we analyze enterprise−level migration strategies, timelines, and broader implications, showing that a comprehensive PQC transition may require 5–15 years. The analysis highlights crypto−agility and effective cross−institutional and cross−jurisdictional coordination as key factors for successful migration, providing insights into the evolution of financial infrastructures and digital trust systems in the post−quantum era.

The migration of post−quantum cryptographic protocols is a core part of the overall transition to post−quantum cryptography. Its main task is to replace the key exchange, identity authentication, digital signature, and other basic cryptographic algorithms used in existing security protocols, such as TLS (transport layer security), IPsec (Internet protocol security), and SSH (secure shell), with post−quantum algorithms, while maintaining system stability and compatibility with deployed protocol versions. This paper investigates heterogeneous authenticated key exchange (AKE) in migration scenarios where only one communicating party has completed the transition to post−quantum cryptography. First, typical homogeneous and heterogeneous AKE constructions and classical authenticated key exchange mechanisms in mainstream security protocols are reviewed. Then, based on the TLS 1.3 mutual authentication framework, heterogeneous authentication combinations are implemented, in which classical RSA and SM2 signature authentication and post−quantum Dilithium signatures (SIG) and Kyber key encapsulation mechanisms (KEM) are used in different authentication directions. The certificate, key loading, handshake authentication, and key derivation processes are extended accordingly. Experiments on eight heterogeneous combinations show that heterogeneous AKE supports combined classical and post−quantum authentication without requiring both parties to upgrade simultaneously, providing a practical transition approach for gradual and agile protocol migration.

Quantum computing erodes the security foundation of classical public−key systems, prompting global financial market infrastructures (FMIs) to accelerate their migration to post−quantum cryptography (PQC). The Phase 2 empirical tests of the Bank for International Settlements (BIS) Project Leap reveal the engineering constraints of shifting real−time gross settlement (RTGS) systems to hybrid cryptography: native schemes do not support parallel processing of heterogeneous algorithms, and end−to−end PQC signature verification takes about 209.9 ms, which is 7.5 times the 28.1 ms of classical cryptography. Our analysis shows that this overhead stems mainly from system−level stages such as message parsing and verification−software orchestration rather than from the lattice operations themselves; placing it on the synchronous critical path of high−concurrency settlement would depress throughput. This study therefore decouples the processing chain so that the upgrade can proceed in stages: a nested signature with an inner PQC layer and an outer RSA layer is constructed in the Business Application Header (BAH), allowing legacy gateways to accept messages without parsing PQC extension blocks (zero−intrusion access); the verification latency is offset by the asynchronous rollback of a distributed transaction, in which a pre−settlement hold removes synchronous blocking and a compensating transaction restores global consistency, overlaid with a quantified monitoring model featuring anomaly−triggered circuit breaking. Simulations show that, once verification is moved off the critical path, the settlement−decision latency drops to the millisecond level and the software−only path already meets the conventional 500 ms monitoring ceiling; with an optional FPGA pipeline accelerator to suppress jitter, the threshold can converge to within 50 ms. The analysis of a three−stage evolution path further supports this conclusion: with proper decoupling and sound asynchronous orchestration, real−time throughput need not be sacrificed, and financial core systems can steadily complete their quantum−resistant security upgrade.

Cryptographic module identification is the starting point of post−quantum cryptographic migration in the banking and financial industry, and its results form the basis for compiling a cryptographic asset inventory and prioritizing migration. In view of the large scale, language diversity, deeply nested wrapper layers and complex dependency chains of banking information systems and operating system source code, an LLM−based cryptographic module identification tool is investigated in this paper. Taking the whole codebase and its dependency information as input, the tool completes identification through the collaboration of five agents responsible for data ingestion and preprocessing, call graph construction, cryptographic semantic identification, trustworthy reasoning and reflection, and result aggregation and visualization, and outputs evidence units traceable to source statements. The tool is implemented on the open−source Qwen3−8B model without parameter fine−tuning. On 400 file−level samples constructed from the source code of the Anolis OS kernel, it achieves a precision of 97.50% and a recall of 96.06%. In a comparative experiment against an open−source pattern−matching cryptographic discovery tool, the proposed tool identifies 76 of the 84 genuine cryptographic modules missed by the reference tool; ablation experiments show that the combination of call graph construction and trustworthy reasoning and reflection raises recall from 20.69% to 96.06%. These results indicate that semantic−level criteria extend the scope of cryptographic module identification from implementations explicitly visible in the code text to system−level code organized through registration and encapsulation.

To address the critical challenges faced by core online banking systems in the financial industry during post−quantum cryptographic migration—including business continuity risks, poor compatibility across massive terminal devices, and high−concurrency performance bottlenecks—this paper proposes and implements a full−stack quantum−resistant cryptographic migration architecture for large commercial banks. The architecture adopts a dual−signature fusion mechanism that combines a hybrid TLS (transport layer security) secure channel with the Chinese national cryptographic standard SM2 and the ML−DSA (module−lattice−based digital signature algorithm). This design establishes an end−to−end evolutionary path spanning from underlying hardware terminals to network transmission and further to backend cluster signature verification. Gray−scale high−fidelity validation conducted in a real online banking environment of a commercial bank demonstrates that, under a concurrent transaction pressure of 1000 TPS (transactions per second), the hybrid secure channel establishment response time is optimized to 3.5 ms, and the single signature verification latency is approximately 13.6 ms. This empirical study preliminarily validates the feasibility of high−dimensional lattice−based cryptographic algorithms under the test environment and workload described herein, providing a replicable engineering demonstration for the smooth post−quantum evolution of critical information infrastructure in China's banking sector.

The rapid advancement of quantum computing poses fundamental and systemic threats to the public−key cryptography underpinning financial market infrastructures. Although regulatory authorities across jurisdictions have issued policy frameworks for post−quantum cryptography (PQC) migration, practical progress remains slow. This paper argues that the bottleneck is not algorithm choice: the standardized PQC algorithms and their parameter sets are few in number and are effectively fixed by standards and regulatory policy, leaving little decision space. The real bottleneck is the asymmetry between the scale of legacy systems and the supply of cryptographic expertise. A large financial institution may contain tens of thousands of cryptographic call sites scattered across hundreds of heterogeneous legacy systems and vendor−delivered code, while engineers competent in PQC remain scarce, making it unrealistic to staff every legacy unit with such experts. Accordingly, this paper redefines the role of AI agents in migration as knowledge delivery to legacy system owners who lack a PQC background: discovering cryptographic call sites in code, configuration and third−party components; determining their business semantics and applicable compliance requirements; generating adaptation recommendations and remediation plans traceable to normative provisions; and leaving decision authority with human reviewers. Four capabilities are proposed − discovery, determination, generation and verification − together with an autonomy grading scheme. Five failure modes specific to this positioning are analyzed: missed migration points, attribute misjudgment, non−compliant recommendations, stale rule bases and over−trust. Using an open−source system with financial characteristics as a case, the paper further examines the actual scale and difficulty of migration point discovery, and presents an evaluation design that requires no production environment.

The iterative breakthroughs in quantum computing pose a realistic security threat to public−key cryptosystems built upon traditional mathematical hard problems. As an important field of the national critical information infrastructure, the financial industry relies on the secure and stable operation of its information systems, which bears directly on national economic security and global financial stability. Against the backdrop of major economies actively advancing post−quantum cryptography (PQC) migration, China's financial industry is systematically promoting PQC migration at multiple levels, including the legal and compliance framework, algorithm standardization, key technology research, and regional engineering practice, thereby exploring a practicable path for the financial system to meet quantum security challenges. In the future, building an autonomous, controllable, secure, and efficient quantum−safe protection system for finance through the coordinated efforts of national strategic guidance, rule of law, and industrial practice will be of major strategic significance for consolidating the security foundation of financial infrastructure in the digital economy era and for safeguarding the long−term sound development of the financial industry.

In recent years, low earth orbit (LEO) satellite constellations, represented by Starlink, OneWeb, China Satellite Network Group, and SpaceSail (Qianfan), have developed rapidly and are carrying an increasing volume of Internet traffic. However, the high mobility of LEO satellites causes substantial variations in link capacity, latency, and packet loss, posing significant challenges to end−to−end TCP transport. Existing end−to−end transport optimization approaches for satellite networks either require modifications and upgrades to end host kernels, making large−scale deployment difficult, or are designed primarily for conventional geostationary earth orbit (GEO) satellites and therefore cannot readily adapt to highly dynamic and time−varying LEO environments. To overcome these limitations, this work investigates TCP performance enhancement in LEO satellite networks and proposes StarPEP, a reconfiguration−aware and transparent performance enhancing proxy (PEP). StarPEP proactively regulates the sender's transmission by adjusting the TCP receive window at the proxy. By jointly exploiting the periodic reconfiguration characteristics of LEO satellite networks and dynamic variations in round−trip time (RTT) and bottleneck bandwidth, StarPEP effectively improves TCP transport performance without requiring any modification to end hosts. We implement a prototype of StarPEP on Linux. Trace−driven emulation experiments based on real−world LEO satellite network measurements show that, in terms of the trade−off between throughput and delay, StarPEP achieves closer to the Pareto frontier without modifying end hosts. Compared with the classic PEP using Hybla as its congestion control algorithm, StarPEP reduces end−to−end latency by 94.8%~95.9% while improving throughput by 49.3%~214.0%. Compared with the classic PEP using LeoCC, StarPEP reduces end−to−end latency by 89.1%~92.6% at the cost of only a small decrease in throughput.

Addressing the issues of premature aging and failure in water−based coatings on electric multiple units within the industrial atmospheric environments of northern coastal regions—specifically the limitations of traditional, single−factor accelerated testing in terms of evaluation accuracy and the lack of multi−factor coupled equivalent accelerated evaluation methods—this study utilizes the C4−grade (highly corrosive) polluted marine atmosphere of Qingdao as a representative scenario. Through environmental factor monitoring, key corrosive elements were identified; subsequently, based on the principles of total enthalpy equivalence and pollutant ion conversion, a coupled environmental spectrum accelerated test—integrating salt spray, humid heat, drying, and UV exposure—was designed. The study systematically evaluated coating properties such as thickness, gloss, color difference, adhesion, and electrochemical impedance. By integrating Pearson correlation analysis, a multi−indicator comprehensive index model for service life prediction was constructed to elucidate coating aging mechanisms and forecast service life. During the accelerated testing, the coatings exhibited distinct evolutionary trends characterized by a decline in gloss, a gradual degradation of adhesion, and a slow decrease in low−frequency impedance. The developed linear comprehensive index model demonstrated a relative prediction error of ≤10%, forecasting a service life of 10 to 13 years for the exterior air−drying coatings on EMUs operating in coastal regions. The equivalent accelerated evaluation methodology established in this study effectively balances both acceleration efficiency and correlation accuracy, thereby providing robust technical support for coating selection, process optimization, and full−lifecycle management of EMU coatings in coastal atmospheric environments.

Artificial intelligence (AI) is increasingly becoming a key technological force driving the intelligent transformation of urban space generation and design. This paper systematically reviews research progress on the application of AI technologies from three dimensions: the identification and measurement of urban spatial elements, urban spatial planning and design, and urban spatial operation and management. Specifically, research on the identification and measurement of urban spatial elements mainly involves element perception and quantification, spatial function identification, and pattern mining; research on urban spatial planning and design focuses on support for urban design as well as the identification, evaluation, allocation, and optimization of land use; and research on urban spatial operation and management primarily covers real-time monitoring and the optimization of operations and management in areas such as transportation, energy, safety, and public services. On this basis, the paper further outlines future research directions from three perspectives: adhering to the goal of improving the human settlement environment and promoting sustainable urban development, developing new modes of AI-driven technological integration and human–AI collaboration, and improving regulatory frameworks for AI-enabled urban space generation. The study aims to provide a reference for the application of AI in urban space generation and planning practice.

In recent years, driven by innovation in stem cell theory and technological breakthroughs, relying on the rapid development of engineering technologies such as microfluidics and 3D bioprinting, significant progress has been made in the field of organoid research. The filed of organoid is gradually completing the transformation process from laboratory research to industrialization, penetration into the application markets of biological models, organ transplantation, and biological computing hardware. This paper systematically reviews the global progress in organoid research and development, accurately assessing the current stage of organoid research and industrialization. The study finds that organoid construction has achieved dual breakthroughs in biomimicry and functionality, unlocking new application scenarios, while the industrial system has begun to take shape. On this basis, the study identifies the critical technical bottlenecks and ethical regulatory challenges that urgently need to be addressed in the field of organoid. It further outlines future development directions, including prioritizing new strategies for vascularization and microenvironment simulation, developing novel technologies for in situ monitoring of organoid culture, and formulating supporting ethical regulatory policies and standards. Furthermore, the paper analyzes the current state of organoid development in China from the perspective of three dimensions as the policy support system, the depth of fundamental research, and the competitiveness of industrial chain. Targeted recommendations for high−quality development of organoid in China are then proposed, including: establishing organoid key R&D program, conducting application pilot and demonstrations, improving the standardization system, and designing a tiered governance framework, which aims to provide a reference for leading China toward new heights in the field of organoid.

Huang Danian was an outstanding geophysicist and strategic scientist in China, posthumously awarded the national honorary title of "People's Educator". He gave up a privileged life overseas to return to his motherland. Upholding the lifelong conviction that "Revitalizing China is the duty of our generation", he embraced the greater-good mindset and shouldered the mission of building a powerful nation through science and education. Leading his research team, he devoted himself to deep-earth exploration. He spearheaded breakthroughs in multiple core technologies long blocked by foreign countries, secured major advances in China's key equipment for deep-earth exploration, and laid a solid technical and talent foundation for technological self-reliance and control. He spared no effort in national resource exploration, national defense security and the training of innovative geoscience talents, continuing to work until the very last moment of his life.