Science,  8 OCT 2026, VOL 394, ISSUE 6820

《科学》2026年10月8日，第394卷，6820期

物理学Physics

High-fidelity entangling gates and nonlocal circuits with neutral atoms

中性原子高保真度纠缠门与非局域线路

▲ 作者：SIMON J. EVERED, MUQING XU, SOPHIE H. LI, ALEXANDRA A. GEIM, J. PABLO BONILLA ATAIDES, MARCIN KALINOWSKI, ET AL.

▲链接：

https://www.science.org/doi/10.1126/science.aei1312

▲摘要：在量子信息系统中，低误差纠缠态的制备与操控至关重要。实际应用里，双量子比特纠缠门是主要误差来源，制约着容错架构中的线路深度与性能。

研究组基于中性原子量子处理器，实现了具有高拉比频率平滑振幅脉冲的纠缠受控Z门；借助态选择性读出与量子比特复用完成快速校准，实现了为99.854（4）%的保真度。经损耗后选择性处理，保真度可提升至99.941（3）%，性能可稳定维持10小时。

随后，研究组将这类低误差门应用于相干原子重排的量子线路。通过创建和解纠缠簇态完成性能基准测试，并进一步利用具备长程连接特性的乱序线路研究非局域纠缠态。该方案为实现大深度线路、高效容错量子计算提供了可行路径。

▲ Abstract：The generation and manipulation of entanglement with low error are essential in quantum information systems. In practice, two-qubit entangling gates constitute a dominant error source, limiting circuit depths and performance in fault-tolerant architectures. Using a neutral-atom quantum processor, we realized entangling controlled-Z gates with a high-Rabi-frequency smooth-amplitude pulse, employing state-selective readout and qubit reuse for fast calibration, and achieved a fidelity of 99.854(4)%, which improved to 99.941(3)% upon loss postselection, with stable performance for 10 hours. We then used these low-error gates in quantum circuits with coherent atom rearrangement. Performance was benchmarked by creating and disentangling cluster states, and subsequently, we studied nonlocally entangled states with scrambling circuits featuring longer-range connectivity. Our approach provides a route toward deep-circuit, efficient fault-tolerant quantum computation.

High-capacity associative memory in a quantum-optical spin glass

量子光学自旋玻璃中的高容量联想记忆

▲ 作者：BRENDAN P. MARSH, DAVID ATRI SCHULLER, YUNPENG JI, HENRY S. HUNT, SURYA GANGULI, SARANG GOPALAKRISHNAN, ET AL.

▲链接：

https://www.science.org/doi/10.1126/science.aec3917

▲摘要：霍普菲尔德神经网络通过全对全耦合自旋存储记忆，并依靠平衡态动力学完成记忆提取。存储过多会阻碍记忆提取，随着网络转变为自旋玻璃，阻挫效应会催生数量呈指数级增长的伪模式。

尽管存在这一局限，研究组表明，在量子光学非平衡动力学机制下，伪模式可转化为可靠记忆，记忆提取能力由此得以恢复甚至提升。研究组在由原子与光子构成的受驱耗散自旋玻璃体系中，实验观测到了具有高存储容量的联想记忆。

在一个16自旋网络中，该体系的存储容量比赫布学习规则下霍普菲尔德模型高出7倍。原子运动能够动态调整连接以提升存储容量，其效果类似神经网络中的短时突触可塑性，从而在量子光学系统中实现了学习行为的雏形。

▲ Abstract：The Hopfield neural network stores memories using all-to-all-coupled spins and recalls those memories through equilibrium dynamics. Storing too many hampers recall because frustration causes an exponential number of spurious patterns to arise as the network becomes a spin glass. Despite this, we show that memory recall can be restored, and even enhanced, under quantum-optical nonequilibrium dynamics that convert spurious patterns into reliable memories. We experimentally observe associative memory with high storage capacity in a driven-dissipative spin glass made of atoms and photons. The capacity surpasses that of the Hopfield model under Hebbian learning by up to sevenfold in a 16-spin network. Atomic motion boosts capacity by dynamically modifying connectivity akin to short-term synaptic plasticity in neural networks, realizing a precursor to learning in a quantum-optical system.

人工智能Artificial Intelligence

Advancing mathematics research with AI-driven formal proof search

AI驱动的形式化证明搜索推动数学研究

▲ 作者：GEORGE TSOUKALAS, ANTON KOVSHAROV, SERGEY SHIROBOKOV, ANJA SURINA, MORITZ FIRSCHING, GERGELY BéRCZI, ET AL.

▲链接：

https://www.science.org/doi/10.1126/science.aej2213

▲ 摘要：大语言模型（LLM）在数学任务上的能力持续提升，但模型输出的不可靠性限制了其在数学研究中的应用。一种解决思路是利用LLM生成Lean等语言进行形式化证明，由编译器对每个证明步骤进行核验。

研究组首次证实该方法可规模化解决数学开放难题。他们搭建了面向形式化证明搜索的智能体，自主攻克了353个埃尔德什公开问题中的9个，证明了《整数序列在线百科全书》492条猜想中的44条，还将该智能体部署在组合数学、优化、图论、代数几何与量子光学研究中。

即便是仅交替执行基于LLM生成与Lean核验的基础智能体，也复现了上述埃尔德什问题的求解结果。上述结果表明，形式化证明搜索在助力自主数学发现方面具有很大前景。

▲ Abstract：Large language models (LLMs) increasingly excel at mathematics tasks, but their unreliability limits their utility in mathematics research. A mitigation is to use LLMs to generate formal proofs in languages such as Lean, in which the compiler verifies every proof step. We present the first demonstration of this method’s value in solving open problems at scale. We built an artificial intelligence agent for formal proof search that autonomously resolved nine of 353 open Erd?s problems, proved 44/492 On-Line Encyclopedia of Integer Sequences conjectures, and is being deployed in combinatorics, optimization, graph theory, algebraic geometry, and quantum optics research. Even a basic agent alternating LLM-based generation with Lean-based verification replicated the Erd?s successes. These findings demonstrate the power of formal proof search as an enabler of autonomous mathematical discovery.

化学Chemistry

Asymmetric enzymatic hydrophosphorylation through O₂ activation

氧活化介导的不对称酶促氢膦酰化

▲ 作者：YI ZHOU, YIFEI GE, WESLEY HARRISON AND HUIMIN ZHAO

▲链接：

https://www.science.org/doi/10.1126/science.aef3001

▲摘要：自然界中由酶催化形成碳-磷键的反应极为罕见，这限制了学界利用生物催化手段制备在药物与农用化学品领域应用广泛的含磷化合物。

研究组报道了一种功能重构黄素依赖型酶，借助氧活化实现不对称酶促氢膦酰化反应。机理研究表明，活性氧物种可转化为具有反应活性的磷中心自由基，随后发生自由基加成与酶促氢原子转移，实现了高对映选择性。

该酶可兼容多种在化学催化体系中很难反应的磷氢供体，从而实现高价值含磷骨架的生物合成。该研究将生物催化的适用范围拓展至可编程的碳-磷键构建，并建立了在酶内调控氧反应活性的新模式。

▲ Abstract：Enzymatic carbon-phosphorus bond formation is extremely rare in nature, limiting biocatalytic access to phosphorus-containing compounds that are widely used in pharmaceuticals and agrochemicals. Here, we report an asymmetric enzymatic hydrophosphorylation through oxygen activation using a repurposed flavin-dependent enzyme. Mechanistic studies revealed that reactive oxygen species are converted into productive phosphorus-centered radicals, followed by radical addition and enzymatic hydrogen atom transfer, achieving high enantioselectivity. The enzyme accommodates diverse phosphorus-hydrogen donors that pose challenges to chemical catalysis, enabling the biosynthesis of valuable phosphorus-containing scaffolds. This work expands the scope of biocatalysis to programmable carbon-phosphorus bond formation and establishes a paradigm for channeling oxygen reactivity in enzymes.

地球科学Earth Science

Insights from more than a year of ongoing episodic lava fountaining at Kīlauea

基拉韦厄火山持续一年多幕式熔岩喷泉喷发的新认识

▲ 作者：A. F. FLINDERS, N. I. DELIGNE, K. HON, D. T. DOWNS, K. J. LYNN, P. A. NADEAU, ET AL.

▲链接：

https://www.science.org/doi/10.1126/science.aef2931

▲摘要：幕式熔岩喷泉喷发在全球范围内十分罕见，自1823年以来，基拉韦厄火山（夏威夷）仅发生过3次。2024年12月起，哈雷茂茂山顶破火山口内发生的一次喷发，截至目前已引发了54次熔岩喷泉事件，每次喷发通常持续数小时，活动间隔为数天至数周。

学界虽已提出多种作用机制，但关于岩浆储库压力、通道渗透率与挥发分出溶作用如何控制熔岩喷泉喷发等核心问题仍有待解决。

研究组依托现代地球化学、地球物理与地质监测手段，对基拉韦厄火山幕式熔岩喷泉进行多参数观测。观测结果揭示，熔岩喷泉与山顶岩体膨胀存在系统性关联，据此可实现前所未有的定量预测。

此次喷发将为理解全球幕式熔岩喷泉的发生时间、演化过程与驱动机制提供新的约束条件，同时凸显基拉韦厄火山作为火山过程天然实验室的重要价值。

▲ Abstract：Episodic lava fountaining is rare globally, with only three occurrences from Kīlauea (Hawaii) since 1823. Beginning in December 2024, an eruption within the summit crater, Halema?uma?u, has produced 54 fountaining episodes to date, each lasting typically for hours and separated by days to weeks. Although multiple mechanisms have been proposed, fundamental questions remain about how magma reservoir pressure, conduit permeability, and volatile exsolution control fountaining onset. We present a multiparameter record of episodic fountaining at Kīlauea using modern geochemical, geophysical, and geologic monitoring. These observations revealed systematic correlations between fountaining and summit inflation, enabling unprecedented quantitative forecasts. This eruption will provide new constraints on the timing, evolution, and processes driving episodic fountaining globally, and highlights Kīlauea’s role as a natural laboratory for studying volcanic processes.

Recent strengthening of eastern Pacific ENSO in the last millennium paleorecord

千年古气候记录中赤道东太平洋厄尔尼诺-南方涛动的近期增强

▲ 作者：J. E. COLE, D. M. THOMPSON, K. A. DYEZ, C. J. TRIPP, A. W. TUDHOPE, M. LOFVERSTROM, ET AL.

▲链接：

https://www.science.org/doi/10.1126/science.ady2660

▲ 摘要：太平洋厄尔尼诺-南方涛动（ENSO）引发了各类极端气候事件，威胁全球生态系统、基础设施与人类福祉。受观测数据匮乏以及气候模型不确定性的双重影响，该气候系统对全球变暖的响应机制缺乏明确约束。

过去千年加拉帕戈斯群岛珊瑚骨骼的地球化学记录显示，与现有古气候记录相比，赤道东太平洋海表温度的年际变率近期大幅增强，超出模拟的自然变率范围。这一变率上升与全球升温同步，是厄尔尼诺现象加剧的结果。

中太平洋珊瑚数据也显示海表温度变率增加，只是信号不如加拉帕戈斯群岛显著。该研究结果为理解ENSO的变率趋势提供了长期背景，也预示极端气候形势不容乐观。

▲ Abstract：The Pacific El Ni?o–Southern Oscillation (ENSO) generates climate extremes that endanger ecosystems, infrastructure, and human well-being worldwide. The response of this system to climate warming is poorly constrained, owing to both data scarcity and uncertainties in climate models. The geochemistry of Galápagos coral skeletons across the past millennium reveals a large recent increase in interannual variability of sea surface temperature in the eastern equatorial Pacific compared with that in existing paleorecords that exceeds simulated natural variability. This increase parallels rising global temperature and results from stronger El Ni?o events. Central Pacific coral data also show increased variability, although less distinctly than in the Galápagos. Our results provide long-term context for understanding ENSO variability trends, with troubling implications for climate extremes.