《自然》（20240118出版）一周论文导读

 

Nature, 18 January 2024, VOL 625, ISSUE 7995

《自然》2024年1月18日，第625卷，7995期

天文学Astronomy

Ionized gas extends over 40 kpc in an odd radio circle host galaxy

电离气体在奇异射电圈宿主星系中延伸超过40kpc

▲ 作者：Alison L. Coil, Serena Perrotta, David S. N. Rupke, Cassandra Lochhaas, Christy A. Tremonti, Aleks Diamond-Stanic, et al.

▲ 链接：

https://www.nature.com/articles/s41586-023-06752-8

▲ 摘要：

2021年发现的一类新河外天文源被命名为奇异射电圈（ORCs），是在天空中跨度约1弧分的微弱弥漫射电连续辐射的大环。几个ORC中心的星系光度红移为z?0.3-0.6，这意味着射电辐射的物理范围直径为几百kpc，但其来源尚不清楚。

研究组报道了一个ORC的光谱数据，其中包括ORC4中心星系在z = 0.4512处的强[O II]发射追踪电离气体。[O II]发射的物理范围直径约为40kpc，大于典型早期星系所预期，但比大规模射电连续体发射小一个数量级。

研究还探测到横跨[O II]星云、大约200千米/秒的速度梯度，以及大约180千米/秒的高速色散。[O II]的等效宽度（大约50?）对于一个静止星系而言是非常高的。

[O II]发射的形态、运动学和强度与星系附近激波电离气体的流入一致，之后是一个更大的、向外移动的激波。延伸的光学和射电发射虽然观测尺度上存在很大差异，但可能是由同一戏剧性事件引发的。

▲ Abstract：

A new class of extragalactic astronomical sources discovered in 2021, named odd radio circles (ORCs), are large rings of faint, diffuse radio continuum emission spanning approximately 1arcminute on the sky. Galaxies at the centres of several ORCs have photometric redshifts of z?0.3–0.6, implying physical scales of several 100 kpc in diameter for the radio emission, the origin of which is unknown. Here we report spectroscopic data on an ORC including strong [O II] emission tracing ionized gas in the central galaxy of ORC4 at z =0.4512. The physical extent of the [O II] emission is approximately 40 kpc in diameter, larger than expected for a typical early-type galaxy but an order of magnitude smaller than the large-scale radio continuum emission. We detect an approximately 200kms1 velocity gradient across the [O II] nebula, as well as a high velocity dispersion of approximately 180 kms1. The [O II] equivalent width (approximately 50?) is extremely high for a quiescent galaxy. The morphology, kinematics and strength of the [OII] emission are consistent with the infall of shock ionized gas near the galaxy, following a larger, outward-moving shock. Both the extended optical and radio emission, although observed on very different scales, may therefore result from the same dramatic event.

人工智能Artificial Intelligence

Solving olympiad geometry without human demonstrations

无需人工演示即可解决奥数几何问题

▲ 作者：Trieu H. Trinh, Yuhuai Wu, Quoc V. Le, He He & Thang Luong

▲ 链接：

https://www.nature.com/articles/s41586-023-06747-5

▲ 摘要：

在奥林匹克竞赛中证明数学定理代表了人类水平自动推理的一个显著里程碑，因为其高难度在世界顶尖大学预科数学人才中是公认的。然而，当前机器学习方法并不适用于大多数数学领域，因为将人类演示翻译为机器可验证格式的成本很高。由于几何具有独特的翻译挑战，从而导致训练数据严重稀缺，因此该问题更加严重。

研究组提出了AlphaGeometry，这是一个欧几里得平面几何的定理证明器，通过综合数百万个不同复杂程度的定理和证明，避开了人类演示的需要。AlphaGeometry是一个神经符号系统，它使用神经语言模型，在该大规模合成数据上从零开始训练，指导符号演绎引擎在颇具挑战性的问题中通过无限分支点。

在一组由30个最新奥数问题组成的测试集中，AlphaGeometry解决了25个问题，优于之前只解决10个问题的最佳方法，接近国际数学奥林匹克（IMO）金牌得主的平均水平。值得注意的是，AlphaGeometry生成了人类可读的证明，在人类专家的评估下解决了IMO 2000年和2015年的所有几何问题，并在2004年发现了IMO定理转化的广义版本。

▲ Abstract：

Proving mathematical theorems at the olympiad level represents a notable milestone in human-level automated reasoning, owing to their reputed difficulty among the world’s best talents in pre-university mathematics. Current machine-learning approaches, however, are not applicable to most mathematical domains owing to the high cost of translating human proofs into machine-verifiable format. The problem is even worse for geometry because of its unique translation challenges, resulting in severe scarcity of training data. We propose AlphaGeometry, a theorem prover for Euclidean plane geometry that sidesteps the need for human demonstrations by synthesizing millions of theorems and proofs across different levels of complexity. AlphaGeometry is a neuro-symbolic system that uses a neural language model, trained from scratch on our large-scale synthetic data, to guide a symbolic deduction engine through infinite branching points in challenging problems. On a test set of 30 latest olympiad-level problems, AlphaGeometry solves 25, outperforming the previous best method that only solves ten problems and approaching the performance of an average International Mathematical Olympiad (IMO) gold medallist. Notably, AlphaGeometry produces human-readable proofs, solves all geometry problems in the IMO 2000 and 2015 under human expert evaluation and discovers a generalized version of a translated IMO theorem in 2004.

材料科学Materials Science

Multifunctional ytterbium oxide buffer for perovskite solar cells

多功能氧化镱缓冲层助力钙钛矿太阳能电池

▲ 作者：Peng Chen, Yun Xiao, Juntao Hu, Shunde Li, Deying Luo, Rui Su, et al.

▲ 链接：

https://www.nature.com/articles/s41586-023-06892-x

▲ 摘要：

钙钛矿太阳能电池（PSCs）包括夹在几层不同电荷选择材料之间的固体钙钛矿吸收体，确保器件的单向电流流动和高压输出。在p型/本征/n型（p-i-n）PSCs（也称为反式PSCs）中，电子选择层和金属电极之间的“缓冲材料”使电子能够从电子选择层流向电极。

此外，这些缓冲材料作为屏障，还可抑制有害物质进入钙钛矿吸收剂或降解产物从吸收剂中流出的相互扩散。到目前为止，可蒸发有机分子和原子层沉积的金属氧化物已获成功，但每种都有特定的缺陷。

研究组报道了一种化学稳定的氧化镱（YbOx）多功能缓冲层，通过可扩展热蒸发沉积制备p-i-n PSCs。在具有窄带隙钙钛矿吸收体的p-i-n PSC中使用这种YbOx缓冲材料，获得了超过25%的认证功率转换效率。

研究组还证明了YbOx在从各种类型的钙钛矿吸收层中实现高效PSC方面的广泛适用性，宽带隙钙钛矿吸收层和中带隙钙钛矿吸收层的最优效率分别为20.1%和22.1%。此外，当遭受ISOS-L-3加速老化时，YbOx封装的器件表现出明显增强的器件稳定性。

▲ Abstract：

Perovskite solar cells (PSCs) comprise a solid perovskite absorber sandwiched between several layers of different charge-selective materials, ensuring unidirectional current flow and high voltage output of the devices. A ‘buffer material’ between the electron-selective layer and the metal electrode in p-type/intrinsic/n-type (p-i-n) PSCs (also known as inverted PSCs) enables electrons to flow from the electron-selective layer to the electrode. Furthermore, it acts as a barrier inhibiting the inter-diffusion of harmful species into or degradation products out of the perovskite absorber. Thus far, evaporable organic molecules and atomic-layer-deposited metal oxides have been successful, but each has specific imperfections. Here we report a chemically stable and multifunctional buffer material, ytterbium oxide (YbOx), for p-i-n PSCs by scalable thermal evaporation deposition. We used this YbOx buffer in the p-i-n PSCs with a narrow-bandgap perovskite absorber, yielding a certified power conversion efficiency of more than 25%. We also demonstrate the broad applicability of YbOx in enabling highly efficient PSCs from various types of perovskite absorber layer, delivering state-of-the-art efficiencies of 20.1% for the wide-bandgap perovskite absorber and 22.1% for the mid-bandgap perovskite absorber, respectively. Moreover, when subjected to ISOS-L-3 accelerated ageing, encapsulated devices with YbOx exhibit markedly enhanced device stability.

化学Chemistry

Computational prediction of complex cationic rearrangement outcomes

复杂阳离子重排结果的计算预测

▲ 作者：Tomasz Klucznik, Leonidas-Dimitrios Syntrivanis, Sebastian Ba?, Barbara Mikulak-Klucznik, Martyna Moskal, Sara Szymku?, et al.

▲ 链接：

https://www.nature.com/articles/s41586-023-06854-3

▲ 摘要：

近年来，人们对计算机辅助有机合成重新产生了兴趣。可规划多步合成途径的反应和神经网络算法的使用已彻底改变了该领域，包括生成先进天然产物的实例。这些方法通常基于完整的、文献推导的“底物到产物”反应规则，无法轻易扩展到反应机理的分析。

研究组展示了配备物理有机化学规则以及量子力学和动力学计算增强的机理步骤的综合知识库计算机，可使用反应网络方法来分析一些最复杂的有机转化机理：即阳离子重排。该重排是有机化学教科书的基石，并导致分子碳骨架发生显著变化。

研究组在https://HopCat.allchemy.net/上描述和部署的算法可在几分钟内生成可能的机理步骤网络，跟踪看似合理的步骤序列并计算预期的产物分布。通过三组实验验证了这一算法，即使对训练有素的化学家而言，这些实验的分析也颇具挑战性：（1）预测尾部至头部萜烯（THT）环化的结果，其中在微小结构细节不同的模块化前体中编码了有实质性区别的结果；（2）比较溶液或超分子胶囊中THT环化的结果；（3）分析复杂反应混合物。

该研究结果支持这样一种愿景，即计算机不再只是操纵已知的反应类型，而是将助力合理化和发现新型且机理复杂的转化。

▲ Abstract：

Recent years have seen revived interest in computer-assisted organic synthesis. The use of reaction- and neural-network algorithms that can plan multistep synthetic pathways have revolutionized this field, including examples leading to advanced natural products. Such methods typically operate on full, literature-derived ‘substrate(s)-to-product’ reaction rules and cannot be easily extended to the analysis of reaction mechanisms. Here we show that computers equipped with a comprehensive knowledge-base of mechanistic steps augmented by physical-organic chemistry rules, as well as quantum mechanical and kinetic calculations, can use a reaction-network approach to analyse the mechanisms of some of the most complex organic transformations: namely, cationic rearrangements. Such rearrangements are a cornerstone of organic chemistry textbooks and entail notable changes in the molecule’s carbon skeleton. The algorithm we describe and deploy at https://HopCat.allchemy.net/ generates, within minutes, networks of possible mechanistic steps, traces plausible step sequences and calculates expected product distributions. We validate this algorithm by three sets of experiments whose analysis would probably prove challenging even to highly trained chemists: (1) predicting the outcomes of tail-to-head terpene (THT) cyclizations in which substantially different outcomes are encoded in modular precursors differing in minute structural details; (2) comparing the outcome of THT cyclizations in solution or in a supramolecular capsule; and (3) analysing complex reaction mixtures. Our results support a vision in which computers no longer just manipulate known reaction types but will help rationalize and discover new, mechanistically complex transformations.

地球科学Earth Science

Ubiquitous acceleration in Greenland Ice Sheet calving from 1985 to 2022

从1985年到2022年格陵兰冰盖崩解普遍加速

▲ 作者：Chad A. Greene, Alex S. Gardner, Michael Wood & Joshua K. Cuzzone

▲ 链接：

https://www.nature.com/articles/s41586-023-06863-2

▲ 摘要：

在过去的几十年里，格陵兰岛几乎所有的冰川都变薄或退缩，导致冰川融化加速，海平面上升速度加快，并对全球气候产生影响。

为了了解冰裂前沿退缩如何影响格陵兰岛的冰质量平衡，研究组结合了1985年至2022年收集的236328次人工推导和人工智能推导的冰川终点位置观测数据，并在近40年里每月生成一个确定冰盖范围的120米分辨率的地图。

结果表明，自1985年以来，格陵兰冰盖（GrIS）已损失了5091±72 km2的面积，对应于1034±120 Gt的冰损失。由于忽略了冰裂前缘退缩，目前对冰盖质量平衡的共识估计低估了格陵兰岛最近的质量损失高达20%。此次报告的冰盖质量损失对全球海平面的直接影响很小，但足以影响全球海洋环流和热能的分布。

在季节性时间尺度上，格陵兰岛每年损失193±25 km2（63±6 Gt）的冰，5月冰损失最大，而9月至10月冰损失最小。研究结果发现，多年代际冰盖退缩与每个冰川的季节性进退幅度高度相关，这意味着季节性时间尺度上的冰盖终点位置变化可作为冰川对长期气候变化敏感性的指标。

▲ Abstract：

Nearly every glacier in Greenland has thinned or retreated over the past few decades, leading to glacier acceleration, increased rates of sea-level rise and climate impacts around the globe. To understand how calving-front retreat has affected the ice-mass balance of Greenland, we combine 236,328 manually derived and AI-derived observations of glacier terminus positions collected from 1985 to 2022 and generate a 120-m-resolution mask defining the ice-sheet extent every month for nearly four decades. Here we show that, since 1985, the Greenland Ice Sheet (GrIS) has lost 5,091 ± 72 km2 of area, corresponding to 1,034 ± 120 Gt of ice lost to retreat. Our results indicate that, by neglecting calving-front retreat, current consensus estimates of ice-sheet mass balance have underestimated recent mass loss from Greenland by as much as 20%. The mass loss we report has had minimal direct impact on global sea level but is sufficient to affect ocean circulation and the distribution of heat energy around the globe. On seasonal timescales, Greenland loses 193 ± 25 km2 (63 ± 6 Gt) of ice to retreat each year from a maximum extent in May to a minimum between September and October. We find that multidecadal retreat is highly correlated with the magnitude of seasonal advance and retreat of each glacier, meaning that terminus-position variability on seasonal timescales can serve as an indicator of glacier sensitivity to longer-term climate change.

社会学Sociology

Online searches to evaluate misinformation can increase its perceived veracity

在线搜索以评估错误信息可提高其感知准确性

▲ 作者：Kevin Aslett, Zeve Sanderson, William Godel, Nathaniel Persily, Jonathan Nagler & Joshua A. Tucker

▲ 链接：

https://www.nature.com/articles/s41586-023-06883-y

▲ 摘要：

学术界对理解网上错误信息的信任度给予了相当大的关注，尤其关注社交网络。然而，尽管使用在线搜索来评估信息真实性是媒介素养干预的核心组成部分，但搜索引擎在信息环境中的主导作用仍未得到充分探索。尽管传统观点认为，在评估错误信息时上网搜索会降低人们对它的信任度，但几乎没有实验证据来评估这一说法。

通过五个实验，研究组提出了一致的证据，证明在线搜索评估虚假新闻文章的真实性实际上增加了人们相信这些新闻的概率。为了阐明这种关系，他们将调查数据与使用自定义浏览器扩展收集的数字跟踪数据结合起来。结果发现，搜索效果集中在搜索引擎返回低质量信息的个人身上。

该研究结果表明，那些在网上搜索以评估错误信息的人有陷入数据空白或低质量来源确凿证据的信息空间的风险。研究组还发现一致的证据表明，在线搜索以评估新闻会增加人们对低质量来源真实新闻的信任度，但不一致的证据表明，其会增加人们对主流来源真实新闻的信任度。

该研究结果强调，媒体素养计划需要将其建议建立在经过实证检验的策略基础上，搜索引擎需要投资于该文所述挑战的解决方案。

▲ Abstract：

Considerable scholarly attention has been paid to understanding belief in online misinformation, with a particular focus on social networks. However, the dominant role of search engines in the information environment remains underexplored, even though the use of online search to evaluate the veracity of information is a central component of media literacy interventions. Although conventional wisdom suggests that searching online when evaluating misinformation would reduce belief in it, there is little empirical evidence to evaluate this claim. Here, across five experiments, we present consistent evidence that online search to evaluate the truthfulness of false news articles actually increases the probability of believing them. To shed light on this relationship, we combine survey data with digital trace data collected using a custom browser extension. We find that the search effect is concentrated among individuals for whom search engines return lower-quality information. Our results indicate that those who search online to evaluate misinformation risk falling into data voids, or informational spaces in which there is corroborating evidence from low-quality sources. We also find consistent evidence that searching online to evaluate news increases belief in true news from low-quality sources, but inconsistent evidence that it increases belief in true news from mainstream sources. Our findings highlight the need for media literacy programmes to ground their recommendations in empirically tested strategies and for search engines to invest in solutions to the challenges identified here.