Explore the vast range of titles available.

of a model—that is, the invariance of its behavior under meaning-preserving alternations in its input—is a highly desirable property in natural language processing. In this paper we study the question: Are Pretrained Language Models (PLMs) consistent with respect to factual knowledge? To this end, we create 🤘, a high-quality resource of cloze-style query English paraphrases. It contains a total of 328 paraphrases for 38 relations. Using 🤘, we show that the consistency of all PLMs we experiment with is poor— though with high variance between relations. Our analysis of the representational spaces of PLMs suggests that they have a poor structure and are currently not suitable for representing knowledge robustly. Finally, we propose a method for improving model consistency and experimentally demonstrate its effectiveness.

Many 2D few-layer materials show piezoelectric or pyroelectric effects due to the loss-of-inversion symmetry induced by broken structure, although they are not piezoelectric or pyroelectric in the bulk. In this work, we find that the puckered graphene-like 2D few-layer black phosphorene is pyroelectric and shows a pyro-catalytic effect, where the pyroelectric charges generated under ambient cold–hot alternation are utilized for hydrogen evolution and dye molecule decomposition. Under thermal cycling between 15 °C and 65 °C, the 2D few-layer black phosphorene shows a direct hydrogen generation of about 540 μmol per gram of catalyst after 24 thermal cycles and about 99% decomposition of Rhodamine B dye after 5 thermal cycles. This work opens a door for the pyro-catalytic energy harvesting from the cold–hot alternations by a class of 2D few-layer materials. 2D elemental materials, with their atomic-scale dimensions, present exciting opportunities for energy conversion applications. Here, the authors use few-layer black phosphorene to perform pyro-catalysis, in which thermal cycling the material induces hydrogen gas production and dye degradation.

The tumor microenvironment (TME) of nasopharyngeal carcinoma (NPC) harbors a heterogeneous and dynamic stromal population. A comprehensive understanding of this tumor-specific ecosystem is necessary to enhance cancer diagnosis, therapeutics, and prognosis. However, recent advances based on bulk RNA sequencing remain insufficient to construct an in-depth landscape of infiltrating stromal cells in NPC. Here we apply single-cell RNA sequencing to 66,627 cells from 14 patients, integrated with clonotype identification on T and B cells. We identify and characterize five major stromal clusters and 36 distinct subpopulations based on genetic profiling. By comparing with the infiltrating cells in the non-malignant microenvironment, we report highly representative features in the TME, including phenotypic abundance, genetic alternations, immune dynamics, clonal expansion, developmental trajectory, and molecular interactions that profoundly influence patient prognosis and therapeutic outcome. The key findings are further independently validated in two single-cell RNA sequencing cohorts and two bulk RNA-sequencing cohorts. In the present study, we reveal the correlation between NPC-specific characteristics and progression-free survival. Together, these data facilitate the understanding of the stromal landscape and immune dynamics in NPC patients and provides deeper insights into the development of prognostic biomarkers and therapeutic targets in the TME. The tumor microenvironment can influence patient survival response to therapy. Here, the authors used single-cell sequencing to investigate the microenvironment of nasopharyngeal cancer and identify tumor-specific signatures in five stromal clusters of cells that may influence patient survival.

The long-term history of the Earth—Moon system as reconstructed from the geological record remains unclear when based on fossil growth bands and tidal laminations. A possibly more robust method is provided by the sedimentary record of Milankovitch cycles (climatic precession, obliquity, and orbital eccentricity), whose relative ratios in periodicity change over time as a function of a decreasing Earth spin rate and increasing lunar distance. However, for the critical older portion of Earth’s history where information on Earth—Moon dynamics is sparse, suitable sedimentary successions in which these cycles are recorded remain largely unknown, leaving this method unexplored. Here we present results of cyclostratigraphic analysis and high-precision U—Pb zircon dating of the lower Paleoproterozoic Joffre Member of the Brockman Iron Formation, NW Australia, providing evidence for Milankovitch forcing of regular lithological alternations related to Earth’s climatic precession and orbital eccentricity cycles. Combining visual and statistical tools to determine their hierarchical relation, we estimate an astronomical precession frequency of 108.6 ± 8.5 arcsec/y, corresponding to an Earth—Moon distance of 321,800 ± 6,500 km and a daylength of 16.9 ± 0.2 h at 2.46 Ga. With this robust cyclostratigraphic approach, we extend the oldest reliable datum for the lunar recession history by more than 1 billion years and provide a critical reference point for future modeling and geological investigation of Precambrian Earth—Moon system evolution.

In this paper, a new stochastic optimization algorithm is introduced, called Driving Training-Based Optimization (DTBO), which mimics the human activity of driving training. The fundamental inspiration behind the DTBO design is the learning process to drive in the driving school and the training of the driving instructor. DTBO is mathematically modeled in three phases: (1) training by the driving instructor, (2) patterning of students from instructor skills, and (3) practice. The performance of DTBO in optimization is evaluated on a set of 53 standard objective functions of unimodal, high-dimensional multimodal, fixed-dimensional multimodal, and IEEE CEC2017 test functions types. The optimization results show that DTBO has been able to provide appropriate solutions to optimization problems by maintaining a proper balance between exploration and exploitation. The performance quality of DTBO is compared with the results of 11 well-known algorithms. The simulation results show that DTBO performs better compared to 11 competitor algorithms and is more efficient in optimization applications.

Dietary fiber is fermented by the human gut microbiota, producing beneficial microbial metabolites, such as short-chain fatty acids. Over the last few centuries, dietary fiber intake has decreased tremendously, leading to detrimental alternations in the gut microbiota. Such changes in dietary fiber consumption have contributed to the global epidemic of obesity, type 2 diabetes, and other metabolic disorders. The responses of the gut microbiota to the dietary changes are specific to the type, amount, and duration of dietary fiber intake. The intricate interplay between dietary fiber and the gut microbiota may provide clues for optimal intervention strategies for patients with type 2 diabetes and other noncommunicable diseases. In this review, we summarize current evidence regarding dietary fiber intake, gut microbiota modulation, and modification in human health, highlighting the type-specific cutoff thresholds of dietary fiber for gut microbiota and metabolic outcomes.

System-wide metabolic homeostasis is crucial for maintaining physiological functions of living organisms. Stable-isotope tracing metabolomics allows to unravel metabolic activity quantitatively by measuring the isotopically labeled metabolites, but has been largely restricted by coverage. Delineating system-wide metabolic homeostasis at the whole-organism level remains challenging. Here, we develop a global isotope tracing metabolomics technology to measure labeled metabolites with a metabolome-wide coverage. Using Drosophila as an aging model organism, we probe the in vivo tracing kinetics with quantitative information on labeling patterns, extents and rates on a metabolome-wide scale. We curate a system-wide metabolic network to characterize metabolic homeostasis and disclose a system-wide loss of metabolic coordinations that impacts both intra- and inter-tissue metabolic homeostasis significantly during Drosophila aging. Importantly, we reveal an unappreciated metabolic diversion from glycolysis to serine metabolism and purine metabolism as Drosophila aging. The developed technology facilitates a system-level understanding of metabolic regulation in living organisms. Stable-isotope tracing allows quantifying metabolic activity by measuring isotopically labeled metabolites, but its metabolome coverage has been limited. Here, the authors develop a global isotope tracing approach with metabolome-wide coverage and use it to characterize metabolic activities in aging Drosophila .

According to observational evidence and climate model projections, the frequency and intensity of the rapid shift from drought to pluvial (rapid dry‒wet alternation, RDWA) increases as warming intensifies. Given that post‐drought precipitation is a key cause of RDWA, this study focuses on changes in post‐drought precipitation. Climate model projections indicate that the mean post‐drought precipitation will increase by 15.3% during 2071–2100 under the Shared Socio‐economic Pathway (SSP) 585 scenario. The scenario‐averaged response rate of post‐drought precipitation (4.6%/K) to global warming is significantly greater than that of general precipitation (2.7%/K). Furthermore, there will be an increase in the proportion of land area experiencing maximum post‐drought precipitation in autumn. The mean post‐drought extreme precipitation will increase by 20.5% under SSP585, exacerbating the severity of RDWA in a warmer world. The post‐drought thermodynamic (precipitable water) and dynamic (atmospheric vertical velocity) components are both conducive to the future increase in post‐drought precipitation. Plain Language Summary Post‐drought precipitation can directly cause the shift from drought to pluvial (dry‒wet alternation), leading to agricultural losses and natural hazards such as landslides. Therefore, understanding and quantifying how it will change in the future is important. Our results find that mean post‐drought precipitation will increase by about 15.3% and more than half of droughts will be accompanied by extreme precipitation in the following month by the end of the 21st century under high emission scenario. More maximum post‐drought precipitation will occur in autumn, potentially affecting agricultural activities and threatening food security in the future. The response of mean post‐drought precipitation to global warming is stronger than that of general precipitation. Our further analysis suggests that under a warming climate, post‐drought atmospheric conditions are favorable for rainfall formation after droughts. Key Points An increase in post‐drought precipitation will exacerbate the frequency and intensity of rapid dry‐wet alternations The future mean response of post‐drought precipitation to warming (4.5%–4.7%/K) is stronger than that of general precipitation (2.1%–3.1%/K) Increased precipitable water and atmospheric vertical velocity favor the increase in post‐drought precipitation

Exploiting the excellent electronic properties of two-dimensional (2D) materials to fabricate advanced electronic circuits is a major goal for the semiconductor industry 1 , 2 . However, most studies in this field have been limited to the fabrication and characterization of isolated large (more than 1 µm 2 ) devices on unfunctional SiO 2 –Si substrates. Some studies have integrated monolayer graphene on silicon microchips as a large-area (more than 500 µm 2 ) interconnection 3 and as a channel of large transistors (roughly 16.5 µm 2 ) (refs.  4 , 5 ), but in all cases the integration density was low, no computation was demonstrated and manipulating monolayer 2D materials was challenging because native pinholes and cracks during transfer increase variability and reduce yield. Here, we present the fabrication of high-integration-density 2D–CMOS hybrid microchips for memristive applications—CMOS stands for complementary metal–oxide–semiconductor. We transfer a sheet of multilayer hexagonal boron nitride onto the back-end-of-line interconnections of silicon microchips containing CMOS transistors of the 180 nm node, and finalize the circuits by patterning the top electrodes and interconnections. The CMOS transistors provide outstanding control over the currents across the hexagonal boron nitride memristors, which allows us to achieve endurances of roughly 5 million cycles in memristors as small as 0.053 µm 2 . We demonstrate in-memory computation by constructing logic gates, and measure spike-timing dependent plasticity signals that are suitable for the implementation of spiking neural networks. The high performance and the relatively-high technology readiness level achieved represent a notable advance towards the integration of 2D materials in microelectronic products and memristive applications. High-integration-density 2D–CMOS hybrid microchips for memristive applications are made demonstrating in-memory computation and electrical response suitable for the implementation of spiking neural networks representing an advance towards integration of 2D materials in microelectronic products and memristive applications.

An objective assessment of hydrological alterations is crucial for comprehensive water resource management, environmental protection, river ecosystem restoration, and integrated watershed water resource management. To quantitatively assess the ecological hydrologic regime alterations in the Xijiang River Basin in southern China considering the impacts from Longtan Reservoir, daily runoff data from two hydrological stations, Dahuangjiangkou and Wuzhou, are selected from 1973 to 2020. The changes in the flow are analyzed, and the indicators of hydrological alternation/range of variability approach (IHA-RVA) are employed to assess the hydrological regime alternations in the studied basin. The main findings are as follows: (1) Both stations show a decreasing trend in annual runoff, with the moving T-test detecting a change point in the runoff series in 2002. (2) The two stations’ overall hydrologic alteration degrees are 57% and 60%, with a more remarkable variation in the upstream area. Hydrological station changes tend to wane as the distance from the reservoir increases, which indicates that with the increased reservoir-station distance, the impact of the reservoir on the hydrologic process diminishes.