Search

Few-shot named entity recognition (NER) exploits limited annotated instances to identify named mentions. Effectively transferring the internal or external resources thus becomes the key to few-shot NER. While the existing prompt tuning methods have shown remarkable few-shot performances, they still fail to make full use of knowledge. In this work, we investigate the integration of rich knowledge to prompt tuning for stronger few-shot NER. We propose incorporating the deep prompt tuning framework with threefold knowledge (namely TKDP), including the internal 1) context knowledge and the external 2) label knowledge & 3) sememe knowledge. TKDP encodes the three feature sources and incorporates them into soft prompt embeddings, which are further injected into an existing pre-trained language model to facilitate predictions. On five benchmark datasets, the performance of our knowledge-enriched model was boosted by at most 11.53% F1 over the raw deep prompt method, and it significantly outperforms 9 strong-performing baseline systems in 5-/10-/20-shot settings, showing great potential in few-shot NER. Our TKDP framework can be broadly adapted to other few-shot tasks without much effort.

Background: It was essential to identify individuals at high risk of fragility fracture and prevented them due to the significant morbidity, mortality, and economic burden associated with fragility fracture. The quantitative ultrasound (QUS) showed promise in assessing bone structure characteristics and determining the risk of fragility fracture. Aims: To evaluate the performance of a multi-channel residual network (MResNet) based on ultrasonic radiofrequency (RF) signal to discriminate fragility fractures retrospectively in postmenopausal women, and compared it with the traditional parameter of QUS, speed of sound (SOS), and bone mineral density (BMD) acquired with dual X-ray absorptiometry (DXA). Methods: Using QUS, RF signal and SOS were acquired for 246 postmenopausal women. An MResNet was utilized, based on the RF signal, to categorize individuals with an elevated risk of fragility fracture. DXA was employed to obtain BMD at the lumbar, hip, and femoral neck. The fracture history of all adult subjects was gathered. Analyzing the odds ratios (OR) and the area under the receiver operator characteristic curves (AUC) was done to evaluate the effectiveness of various methods in discriminating fragility fracture. Results: Among the 246 postmenopausal women, 170 belonged to the non-fracture group, 50 to the vertebral group, and 26 to the non-vertebral fracture group. MResNet was competent to discriminate any fragility fracture (OR = 2.64; AUC = 0.74), Vertebral fracture (OR = 3.02; AUC = 0.77), and non-vertebral fracture (OR = 2.01; AUC = 0.69). After being modified by clinical covariates, the efficiency of MResNet was further improved to OR = 3.31–4.08, AUC = 0.81–0.83 among all fracture groups, which significantly surpassed QUS-SOS (OR = 1.32–1.36; AUC = 0.60) and DXA-BMD (OR = 1.23–2.94; AUC = 0.63–0.76). Conclusions: This pilot cross-sectional study demonstrates that the MResNet model based on the ultrasonic RF signal shows promising performance in discriminating fragility fractures in postmenopausal women. When incorporating clinical covariates, the efficiency of the modified MResNet is further enhanced, surpassing the performance of QUS-SOS and DXA-BMD in terms of OR and AUC. These findings highlight the potential of the MResNet as a promising approach for fracture risk assessment. Future research should focus on larger and more diverse populations to validate these results and explore its clinical applications.

Recent advancements in blind image quality assessment (BIQA) are primarily propelled by deep learning technologies. While leveraging transformers can effectively capture long-range dependencies and contextual details in images, the significance of local information in image quality assessment can be undervalued. To address this challenging problem, we propose a novel feature enhancement framework tailored for BIQA. Specifically, we devise an Adaptive Graph Attention (AGA) module to simultaneously augment both local and contextual information. It not only refines the post-transformer features into an adaptive graph, facilitating local information enhancement, but also exploits interactions amongst diverse feature channels. The proposed technique can better reduce redundant information introduced during feature updates compared to traditional convolution layers, streamlining the self-updating process for feature maps. Experimental results show that our proposed model outperforms state-of-the-art BIQA models in predicting the perceived quality of images. The code is available at https://github.com/sky-whs/AGAIQA.

In order to solve the problem of transmission power fluctuation and efficiency degradation of wireless power transfer system caused by the change of receiver coil position, this article proposes an omnidirectional wireless power transfer (OWPT) system output power control strategy based on online estimation of maximum mutual inductance parameters. The equivalent circuit model of the OWPT system with three-phase inductance-capacitance-capacitance-serial (LCC-S) compensation network is established, and the bivariate gradient descent method is used to search for the maximum and minimum values of dc input power, thereby estimating the maximum value of mutual inductance parameters. Based on the estimated maximum mutual inductance value and the load receiving power equation, determine the current amplitude of each transmitting coil when the required load receiving power is required. When ensuring the maximum transmission efficiency of the OWPT system, solve the problem of power fluctuations caused by the movement of the receiving coil position in the system. Finally, the feasibility of estimating the maximum mutual inductance parameters and the correctness of the output power control strategy were verified through experiments.

In this paper, a clustered parallel training algorithm is designed for personalized federated learning (Per-FL), called CPPer-FL. CPPer-FL improves the communication and training efficiency of Per-FL from two perspectives, namely, less burden for the central server and lower interaction idling delay. CPPer-FL adopts a client-edge-center learning architecture, which offloads the central server's model aggregation and communication burden to distributed edge servers. Also, CPPer-FL redesigns the cascading model synchronization and updating procedure in conventional Per-FL and changes it to a parallel manner, thus improving the interaction efficiency in the training process. Further, for the proposed hierarchical architecture, two approaches are proposed to cater to Per-FL: similarity-based clustering for client-edge association and personalized model aggregation for parallel model updating, such that clients’ personal features can be preserved in the training process. The convergence of CPPer-FL has been formally analyzed and proved. Evaluation results validate the communication efficiency, model convergence, and model accuracy improvement.

Driving increased consumption of clean energy, the low carbon energy system will be a significant technology to achieve the goal of carbon neutrality. Therefore, the safety and reliability of power equipment in low carbon energy systems is particularly important. The modular multilevel converter (MMC) is widely used in the field of flexible dc transmission and exhibits many advantages. However, in such systems, the enormous number of submodules (SMs) represents potential failure points. To overcome the problem of SM open-circuit for MMCs, a fault diagnosis with data processing of sensors is proposed in this article. The distortion of bridge arm current is used for fault detection, and the prediction of the SM capacitance voltage by whale optimization algorithm-least squares support vector machine (WOA-LSSVM) is used for fault location. All data of characteristic bridge arm current and the SMs capacitance voltage are extracted during normal operation and fault operation. To overcome the defect of parameter randomization of LSSVM neural network, the WOA algorithm is used to optimize the model and further improve its prediction accuracy. In addition, this technique does not require the additional sensors and complex mathematical models. A three-phase MMC prototype with four SMs per arm is built in the laboratory, and the simulation and experimental result confirm the effectiveness of the proposed method.

Data distribution gaps often pose significant challenges to the use of deep segmentation models. However, retraining models for each distribution is expensive and time-consuming. In clinical contexts, device-embedded algorithms and networks, typically unretrainable and unaccessable post-manufacture, exacerbate this issue. Generative translation methods offer a solution to mitigate the gap by transferring data across domains. However, existing methods mainly focus on intensity distributions while ignoring the gaps due to structure disparities. In this paper, we formulate a new image-to-image translation task to reduce structural gaps. We propose a simple, yet powerful Structure-Unbiased Adversarial (SUA) network which accounts for both intensity and structural differences between the training and test sets for segmentation. It consists of a spatial transformation block followed by an intensity distribution rendering module. The spatial transformation block is proposed to reduce the structural gaps between the two images. The intensity distribution rendering module then renders the deformed structure to an image with the target intensity distribution. Experimental results show that the proposed SUA method has the capability to transfer both intensity distribution and structural content between multiple pairs of datasets and is superior to prior arts in closing the gaps for improving segmentation.

The construction of isotypic high-nuclearity inorganic cages with identical pristine parent structure and increasing nuclearity is highly important for molecular growth and structure–property relationship study, yet it still remains a great challenge. Here, we provide an in situ growth approach for successfully synthesizing a series of new giant hollow polymolybdate dodecahedral cages, Mo250, Mo260-I, and Mo260-E, whose structures are growth based on giant polymolybdate cage Mo240. Remarkably, they show two pathways of nuclear growth based on Mo240, that is, the growth of 10 and 20 Mo centers on the inner and outer surfaces to afford Mo250and Mo260-I, respectively, and the growth of 10 Mo centers both on the inner and outer surfaces to give Mo260-E. To the best of our knowledge, this is the first study to display the internal and external nuclear growth of a giant hollow polyoxometalate cage. More importantly, regular variations in structure and nuclearity confer these polymolybdate cages with different optical properties, oxidative activities, and hydrogen atom transfer effect, thus allowing them to exhibit moderate to excellent photocatalytic performance in oxidative cross-coupling reactions between different unactivated alkanes and N-heteroarenes. In particular, Mo240and Mo260-Ewith better comprehensive abilities can offer the desired coupling product with yield up to 92% within 1 h.

Instrument-tissue interaction detection task, which helps understand surgical activities, is vital for constructing computer-assisted surgery systems but with many challenges. Firstly, most models represent instrument-tissue interaction in a coarse-grained way which only focuses on classification and lacks the ability to automatically detect instruments and tissues. Secondly, existing works do not fully consider relations between intra- and inter-frame of instruments and tissues. In the paper, we propose to represent instrument-tissue interaction as $\langle $ instrument class, instrument bounding box, tissue class, tissue bounding box, action class $\rangle $ quintuple and present an Instrument-Tissue Interaction Detection Network (ITIDNet) to detect the quintuple for surgery videos understanding. Specifically, we propose a Snippet Consecutive Feature (SCF) Layer to enhance features by modeling relationships of proposals in the current frame using global context information in the video snippet. We also propose a Spatial Corresponding Attention (SCA) Layer to incorporate features of proposals between adjacent frames through spatial encoding. To reason relationships between instruments and tissues, a Temporal Graph (TG) Layer is proposed with intra-frame connections to exploit relationships between instruments and tissues in the same frame and inter-frame connections to model the temporal information for the same instance. For evaluation, we build a cataract surgery video (PhacoQ) dataset and a cholecystectomy surgery video (CholecQ) dataset. Experimental results demonstrate the promising performance of our model, which outperforms other state-of-the-art models on both datasets.

With the surge of vehicles and transportation, sensing obstacles and warning drivers to avoid accidents have become a great concern in recent years. In the current roadworthy electromagnetic environment, the number of frequency modulated continuous wave (FMCW) millimeter-wave (MMW) automotive radars has exploded due to their unique advantages in environmental sensing. However, the frequency band of the automotive radars is limited from 77 to 81 GHz, hence the burgeoning of radars on the road is bound to cause mutual interference and jeopardize further target detection and parameter estimation. In this paper, two basic schemes are considered to mitigate mutual interference of automotive radars. First, we consider the sparse characteristics of the mutual interference in the two-dimensional (2-D) time domain and employ a sparse interference extraction (SIE) method to tackle the mutual interference. Next, we further consider the low-rank property of the useful echoes across multiple channels and propose a novel three-dimensional (3-D) tensor decomposition (TD) method to decompose the received signals into mutual interference and useful echoes. Several numerical simulations are fulfilled to test the robustness of the proposed TD method, especially for multiple input and multiple output (MIMO) systems under complex electromagnetic circumstances. Furthermore, more experiments are implemented to demonstrate its feasibility in practical applications in comparison to multiple state-of-the-art methods.

ABSTRACTThe surge motion of the platform affects the aerodynamic and structural performance of the floating offshore wind turbine (FOWT) with tip-fusion winglets. Hence, this paper aims to study the aerodynamic and structural performance of the wind turbine with and without tip-fusion winglets under the surge motion by ANSYS software. Comparison of the differences in the pressure distribution, flow fields, modal analysis, and stress of the wind turbines under surge motions. The results show that the tip-fusion winglets have an obvious influence on the pressure distribution and flow fields, and the influence mainly concentrates on the tip of the blade. In addition, the frequency of the wind turbine with fusion winglets is less than without winglets from the first mode to the third mode. The main vibration type of wind turbines with and without winglets is flapping. Besides, the maximum stress of the wind turbine with fusion winglets is increased by 19.91%, and there is a stress concentration phenomenon at the connection between the blade tip and the winglet, which is easy to cause fatigue damage. The location of the maximum stress appears in the transition section between the Cylinder airfoil and the DU airfoil.

• Smart EMI shielding materials are focused and reviewed. • Strain-responsive EMI shielding materials are summarized. • Chemical reagent-responsive EMI shielding materials are summarized. • Shape memory-responsive EMI shielding materials are summarized. • Phase transition-responsive EMI shielding materials are summarized. With the rapid advancement of the intelligent era, intelligent electromagnetic interference (EMI) shielding devices are receiving more and more attention due to their advantages in environmental self-adaption response. Accordingly, appropriate EMI shielding materials are crucial to blocking harmful electromagnetic radiation and passing serviceable electromagnetic waves. Smart EMI shielding materials that can dynamically adjust their EMI shielding effectiveness (SE) in response to specific application requirements and environmental changes are extremely advantageous in both military and civil areas. To date, materials with adjustable EMI SE for various responses have been developed. This review pays special attention to smart materials with tunable EMI SE. The design strategies, mechanism and recent progress of smart EMI shielding materials are discussed together with different stimuli responses, including compression strain, tensile strain, chemical reagent, shape memory, phase transition and crossover angle change-induced responses. The review ends up to discuss challenges and perspectives for smart EMI shielding materials. [Display omitted] [ABSTRACT FROM AUTHOR]

The loss of rare earths (REs) takes place during the pre-decalcification process of mixed rare earth concentrate. In an effort to reduce such RE loss, a novel idea to improve the leaching selectivity of Ca to REs by applying selective mechanical activation was proposed. First, regarding the key minerals affecting the leaching selectivity of Ca to REs, the differences in the mechanical activation behaviors of CaF2and REFCO3were studied, and we find that the lattice strain of CaF2increases from 0.21% to 0.42%, whereas that of REFCO3increases from 0.31% to 0.40%. Notably, CaF2demonstrates a larger lattice strain than REFCO3, indicating greater mechanical activation energy storage and higher leaching activity. Next, the HCl leaching process was studied. A significant leaching selectivity of Ca to REs, from 21.6 to 35.1, is achieved through mechanical activation. The Ca leaching rate reaches 80.7% when the RE loss is 2.3% in the activated sample. This study provides an novel approach for achieving selective extraction of specific components via mechanical activation pretreatment.

Person re-identification aims to retrieve pedestrians with the same identity across different cameras. However, current methods increase attention to interfering regions when dealing with complex backgrounds and occlusion, especially in the presence of similar interfering features. To enhance the robustness of the model, we propose the Reverse Pyramid Attention Guidance (RPAG) network, using a reverse pyramid structure to learn features at multiple granularities. To mitigate the impact of occlusion, we introduce the Similar Feature Filtering (SFF) attention module at the pixel level, using graph convolution to adaptively select occluded regions, thereby enhancing retrieval accuracy by filtering out irrelevant parts. Combining the reverse pyramid structure with the pixel-level attention module strengthens adaptability to complex scenes, guides multi-granularity feature learning, and effectively handles various occlusion scenarios. RPAG achieved Rank-1 accuracies of 96.2%, 93.2%, 88.7%, and 73.2% on the Market1501, DukeMTMC-ReID, MSMT17, and Occluded-Duke datasets, respectively.

The microstructure evolution and strengthening behavior of the ultra-high strength Ti–15Mo–2.7Nb– 3Al–0.2Si titanium alloy were studied utilizing XRD, OM, SEM, and TEM analyses. The results show that the dislocation-strengthening and precipitation-strengthening effects could mostly affect the yield strength of this alloy. The highest yield strength of 1518 MPa was obtained under a combined process of cold rolling + recrystallization + cold rolling + duplex aging. This trend is mainly due to the high density of remaining dislocations, as well as dense and thin secondary αphases in microstructures. A theoretical composite-strengthening model was built, and the prediction error is within 16.6%. Furthermore, it is found that increasing the content of the secondary αphase can continuously strengthen the intragrain zone. This feature causes the intergranular fracture to appear and gradually dominate the fracture surface.

How to identify the ear hole more accurately and more quickly are critical for the active noise reduction in the road vehicle. Obscurations, causing by the long hair, the earphone, and some kind of caps, make the spatial location and orientation of the driver’s ears almost impossible. To solve these two problems, a new method was proposed to replace the original image scaling detection by combining the Haar-like algorithm with the scale size of the variable detection window in the recognition process. A depth positioning camera is added to the ear recognition, which may offer more accurate point vectors for the driver’s two ear holes. A simplified spatial reconstruction model was developed to assist the search when the piercing was occluded. After calculating the image-camera-cab projection mapping relationship, only four feature points are enough for the model’s data requirement. The test results show that the detection Frame Per Second (FPS) was improved to 8.8, and the ear hole’s positioning error is less than 0.1?cm. The proposed system can locate ear holes even with 100% obscurations. The research may have potential application as a front-end subsystem for the active noise reduction and voice interaction systems, improving the quality of acoustic control in cars.

To cope with the increasingly global greenhouse effect and energy shortage, it is urgent to develop a feasible means to convert anthropogenic excess carbon dioxide (CO2) into energy resources. The photocatalytic CO2reduction reaction (CO2RR) coupled with the water oxidation reaction (WOR), known as artificial photosynthesis, is a green, clean, and promoting strategy to deal with the above issues. Among the reported photocatalytic systems for CO2reduction, the main challenge is to achieve WOR simultaneously due to the limited charge separation efficiency and complicated dynamic process. To address the problem, scientists have assembled two nanosemiconductor motifs for CO2RR and WOR into a heterojunction photocatalyst to realize artificial photosynthesis. However, it is difficult to clearly explore the corresponding catalytic mechanism and establish an accurate structure–activity relationship at the molecular level for their aperiodic distribution and complicated structural information. Standing on the shoulders of the heterojunction photocatalysts, a new-generation material, hetero-motif molecular junction (HMMJ) photocatalysts, has been developed and studied by our laboratory. A hetero-motif molecular junction is a class of crystalline materials with a well-defined and periodic structure, adjustable assembly mode, and semiconductor-like properties, which is composed of two predesigned motifs with oxidation and reduction, respectively, by coordination or covalent bonds. The intrinsic properties make these catalysts susceptible to functional modifications to improve light absorption and electrical conductivity. The small size and short distance of the motifs can greatly promote the efficiency of photogenerated electron–hole separation and migration. Based on these advantages, they can be used as potential excellent photocatalysts for artificial photosynthesis. Notably, the explicit structural information determined by single-crystal or powder X-ray diffraction can provide a visual platform to explore the reaction mechanism. More importantly, the connection number, spatial distance, interaction, and arrangement mode of the structural motifs can be well-designed to explore the detailed structure–activity relationship that can be hardly studied in nanoheterojunction photocatalyst systems. In this regard, HMMJ photocatalysts can be a new frontier in artificial photosynthesis and serve as an important bridge between molecular photocatalysts and solid photocatalysts. Thus, it is very important to summarize the state-of-the-art of the HMMJ photocatalysts used for artificial photosynthesis and to give in-depth insight to promote future development.

Polymer-enhanced foam (PEF) can effectively improve the conformance profile and delay the gas breakthrough during gas flooding, but it suffers from low foamability and poor stability under harsh reservoir conditions. This research aims to improve the performance of the PEF by applying a synthesized acrylamide-based quadripolymer, named MACS, that contained active and rigid functional groups. The foamability, thermal stability, blocking performance, and conformance control of MACS-enhanced foam (MEF) were evaluated under varying testing conditions, and the conventional partially hydrolyzed polyacrylamide-enhanced foam (HEF) was used as the control group. Meanwhile, the migration of the MEF in the porous medium and its interaction with the crude oil were also investigated in both 2-D and 3-D models under ambient conditions. Furthermore, the mobility control capability and EOR performance of the MEF and HEF were assessed and compared. The results demonstrated that 0.15 wt % MACS effectively enhanced the durability of the foam in adverse environments (55 °C and 30 000 ppm of NaCl) and the foaming-complexed index (FCI) of MEF was approximately two times greater than that of HEF under identical conditions. MEF performed well in blocking the high-permeability layers and diverting the displacing fluid into the low-permeability region in the 2-D and 3-D models, and relatively uniform displacement fronts were clearly observed. Besides, due to the presence of active functional groups in the MACS, residual oil could be better emulsified and displaced. Moreover, the retention of MACS in the porous medium barely caused any permeability reduction. Finally, core flooding experiments showed that the recovery factor of MEF flooding reached 78.69%, while the counterpart HEF merely recovered 70.27% of the OOIP. In all, this work may provide valuable insights for the development of novel foam boosters, which could find potential uses in reservoirs experiencing difficulties with gas channeling during gas injection.

ABSTRACTIn damaged Fukushima Daiichi NPP reactors, groundwater microorganisms have continuously seeped into contaminated water. We investigated the effect of a siderophore-releasing microorganism (SB) on fuel debris analogs. Fuel debris analog pellet samples (FDAPS) and powder samples containing CeO2(an alternative to UO2)-ZrO2solid solution and metallic iron were formed. FDAPSs were contacted with SB on a membrane filter placed on agar medium for 50 days. With the addition of SB, Fe-containing degradation products were detected by SEM-EDX analyses on FDAPSs, on the filter, and on the agar medium, revealing that some fractions of Fe ions were dissolved and precipitated on FDAPSs, and the rest became detached from FDAPSs and migrated through the filter. The SB-derived degradation differed from those by P. fluorescensand Na-citrate solution. The RBS and ERDA spectrometry analyses identified the degradation products as Fe oxyhydroxides. Although Zr was detected in small amounts in the Fe area by SEM-EDX and by SIMS analyses, the dissolution of Zr may be limited. These results clarify that the presence of SB accelerates the degradation of fuel debris in which Fe metal regions are preferentially degraded, and some fraction of Fe were detached and migrated from the fuel debris analog.

Leaf gas exchange plays a critical role in determining crop final yield, and there is a threshold response of leaf gas exchange to water stress. It is of great significance to quantify crop water stress severity by using the response characteristics of leaf gas exchange to drought. However, it is currently unclear whether leaf gas exchange serve as a reliable indicator for predicting crop final yield in response to drought, which affects the accuracy of monitoring agricultural drought using physiological indicators during the crop growing season. This study determined the response threshold of leaf gas exchange to drought for spring wheat through a serials of soil dry-down experiments and used the threshold characteristics to construct and parameterize a spring wheat growth model. Spring wheat were designed to be irrigated with five treatments (with supplementary irrigation at 230 mm, 165 mm, 115 mm, 50 mm and 0 mm). Crop model were used to simulate and analyze the threshold response characteristics of grain yield to drought and compare them to the thresholds of leaf gas exchange indices for spring wheat. The results showed that the response threshold of stomatal conductance of spring wheat to fraction of transpirable soil water was 0.5, which was greater than that of transpiration rate and net photosynthetic rate, 0.4. The parameterized spring wheat growth model with the response threshold of net photosynthetic rate to fraction of transpirable soil water accurately simulated the aboveground biomass and final yield of spring wheat. The response threshold of spring wheat final yield to fraction of transpirable soil water was significantly smaller than that of leaf gas exchange parameters to fraction of transpirable soil water (0.18 versus 0.4). This indicates that there are certain problems in using physiological indicator such as leaf gas exchange indices during crop growing season to determine the agricultural drought severity and reflect the reduction of final crop yields due to drought.

Cu coatings were prepared via the laser-assisted low-pressure cold spray (LPCS) method to study the effect of laser irradiation on microstructure and corrosion behaviors of the coatings. The results reveal that the laser-assisted LPCS-Cu coatings are denser and have better coating−substrate interfacial bonding as compared to LPCS-Cu coating. Laser irradiation improves the overall plastic deformation of deposited particles, resulting in uniform grain refinement of Cu particles. Therefore, the laser-assisted LPCS-Cu coatings show more uniform microstructure and smaller grain size. The results of electrochemical tests demonstrated that laser-assisted LPCS-Cu coatings have higher corrosion potential, lower corrosion current and corrosion rate than the LPCS-Cu coating in 3.5 wt.% NaCl solution. The main reason is that laser irradiation improves the coating density and intimate bonding between particles, and a continuous and dense corrosion product film composed of CuCl and Cu2O is formed on the surface of the laser-assisted LPCS-Cu coating to block the erosion of the corrosive solution.

Perovskite/perovskite/silicon triple-junction solar cells hold promise for surpassing their two-junction counterparts in performance. Achieving this requires monolithic integration of a ∼2.0 eV band-gap perovskite subcell, characterized by a high bromide:iodide ratio (>7:3), and with low-temperature processability and high optoelectronic quality. However, light-induced phase segregation in such perovskites remains a challenge. To address this, we propose modifying the wide-band-gap perovskite with potassium thiocyanate (KSCN) and methylammonium iodide (MAI) co-additives, where SCN−increases the perovskite grain size, reducing the grain boundary defect density; K+immobilizes the halide, preventing the formation of halide vacancies; and MA+eliminates the residual light-destabilizing SCN−in the perovskite films via double displacement reactions. Our co-additive strategy enables enhanced photostability, whereas individual usage of MAI and KSCN would result in adverse effects. Triple-junction tandem solar cells, incorporating co-additive-modified 2.0 eV perovskites as top cell absorbers, reach a 3.04 V open-circuit voltage and a PCE of 26.4% over a 1 cm2area.

The liquid–liquid interface provides a promising platform to construct supramolecular polymers and materials with advanced functions. However, supramolecular polymerization at the interface usually requires monomers with different or even orthogonal solubilities, which significantly limit the number of usable monomers. Here, we report a new strategy to construct supramolecular polymer networks at the oil–water interface using a water-soluble catechol–iron complex and an oil-soluble diend-functionalized polymer. Owing to the dynamic catechol–iron coordination bond and imine bond, the resulting supramolecular polymer networks demonstrate excellent dynamic features and responsiveness to different stimuli including the pH, redox, competing ligands, and temperature. With a supramolecular polymer network serving as a building block, emulsions and 2D films can be generated that have potential applications in encapsulation, release, and molecular separation.

As a key component of aircraft engines, the complexity of blade features and the increasing production volumes pose significant challenges to the accuracy and efficiency of their profile inspections. In order to concurrently ensure accuracy and efficiency in measuring the maximum thickness of aero-engine blade cross sections, this study presents a measurement method employing bidirectional line-structured light scanning based on blade profile features. The designed bidirectional line structured light scanning system is tailored for measuring blade profile, enabling the simultaneous acquisition of coordinates for all points on both sides of the complete blade profile. A calibration method is introduced to determine the motion direction of the displacement stage within the laser scanning sensor (LSS) coordinate system, as well as the relative position of the bidirectional LSS coordinate system. This method aims to integrate the coordinates of measurement data points acquired into the world coordinate system. In addition, a feature-extracting algorithm, grounded in the inherent profile features of the blade, is proposed to accurately determine the maximum thickness of the blade cross section. The effectiveness of the proposed method is verified through the measurement of an aero-engine blade. A comparison between the maximum blade thickness measured by the proposed method and the results obtained by the coordinate measuring machine (CMM) reveals a maximum difference of 0.003 mm. The proposed method can be extended to other cross-sectional characteristic parameters on the contour of blades, contributing to the broader field of aero-engine design and manufacturing.

Two-terminal monolithic perovskite/silicon tandem solar cells demonstrate huge advantages in power conversion efficiency compared with their respective single-junction counterparts1,2. However, suppressing interfacial recombination at the wide-bandgap perovskite/electron transport layer interface, without compromising its superior charge transport performance, remains a substantial challenge for perovskite/silicon tandem cells3,4. By exploiting the nanoscale discretely distributed lithium fluoride ultrathin layer followed by an additional deposition of diammonium diiodide molecule, we have devised a bilayer-intertwined passivation strategy that combines efficient electron extraction with further suppression of non-radiative recombination. We constructed perovskite/silicon tandem devices on a double-textured Czochralski-based silicon heterojunction cell, which featured a mildly textured front surface and a heavily textured rear surface, leading to simultaneously enhanced photocurrent and uncompromised rear passivation. The resulting perovskite/silicon tandem achieved an independently certified stabilized power conversion efficiency of 33.89%, accompanied by an impressive fill factor of 83.0% and an open-circuit voltage of nearly 1.97 V. To the best of our knowledge, this represents the first reported certified efficiency of a two-junction tandem solar cell exceeding the single-junction Shockley–Queisser limit of 33.7%.

Pancreatic ductal adenocarcinoma (PDAC) is distinguished by its aggressive malignancy, limited treatment avenues and a tendency towards chemotherapy resistance, underscoring the critical need for advanced research to uncover new therapeutic approaches. Stress granules (SGs) that is implicated in cellular self-protection mechanism, along with its associated family molecules have shown pro-cancer effects and are closely related to tumor chemotherapy resistance. In this study we investigated the relationship between Ras GTPase-activating protein-binding proteins 2 (G3BP2), a core component of SGs, and the malignancy of PDAC as well as its resistance to the chemotherapy drug gemcitabine. Analyzing TCGA dataset revealed that the expression of G3BP1 and G3BP2 was significantly upregulated in PDAC compared with adjacent normal pancreatic tissues, and the high expression of G3BP2 rather than G3BP1 was significantly associated with poorer overall survival (OS) in PDAC patients. We demonstrated that knockdown of G3BP2 inhibited the proliferation and invasion of PANC‐1 and CFPAC-1 cells in vitro and in vivo. By analyzing the differentially expressed genes in G3BP2 knockdown and overexpressed PANC‐1 cells, we identified DKC1 that was associated with RNA stability and regulation as the target of G3BP2. We demonstrated that G3BP2 bound to PDIA3 mRNA and recruited them into SGs, increasing the stability of PDIA3 mRNA and attenuating its translation efficiency, thereby promoting DKC1 expression. Furthermore, DKC1 could bind to hENT mRNA and inhibited its expression, which enhanced gemcitabine resistance of PDAC. Therefore, we propose a novel mechanism wherein G3BP2 facilitates PDAC’s resistance to chemotherapy by modulating PDIA3-DKC1-hENT in a SGs-dependent way, suggesting G3BP2 SGs a protentional therapeutic target for the treatment in PDAC.

Emission of volatile organic compounds has important influence on complex air pollution and human health. In this paper, a series of tungsten–iron composite oxides with different proportions and preparation methods were synthesized and first used for catalytic combustion of chlorobenzene and toluene, as typical polluting gas sources. These WO3-based solid catalytic materials were systematically characterized by modern analytical methods, and the results showed that there was strong electron interaction between W and Fe elements in the composite oxides, and the presence of a certain amount of tungsten oxide inhibited the crystallization of iron oxide, and vice versa, which were beneficial to the uniform dispersion of tungsten–iron components into each other and the improvement of redox properties. Compared with single-component oxide, the formation of tungsten–iron composite oxide affected the micro-structure, improved the specific surface area and optimized the pore structure of materials. The performance test results showed that the tungsten–iron composite oxide (FeWO4–0.5Fe2O3, molar ratio of tungsten and iron was 1/2) prepared using citric acid-based sol–gel method was the optimal, and its catalytic degradation efficiency could reach 90% for chlorobenzene and 83% for toluene at 320 °C, and maintain at least 60 h without obvious deactivation, with high selectivity to the formation of HCl and CO2.

Image segmentation is fundamental task for medical image analysis, whose accuracy is improved by the development of neural networks. However, the existing algorithms that achieve high-resolution performance require high-resolution input, resulting in substantial computational expenses and limiting their applicability in the medical field. Several studies have proposed dual-stream learning frameworks incorporating a super-resolution task as auxiliary. In this paper, we rethink these frameworks and reveal that the feature similarity between tasks is insufficient to constrain vessels or lesion segmentation in the medical field, due to their small proportion in the image. To address this issue, we propose a DS2F (Dual-Stream Shared Feature) framework, including a Shared Feature Extraction Module (SFEM). Specifically, we present Multi-Scale Cross Gate (MSCG) utilizing multi-scale features as a novel example of SFEM. Then we define a proxy task and proxy loss to enable the features focus on the targets based on the assumption that a limited set of shared features between tasks is helpful for their performance. Extensive experiments on six publicly available datasets across three different scenarios are conducted to verify the effectiveness of our framework. Furthermore, various ablation studies are conducted to demonstrate the significance of our DS2F.

Flexible tactile sensors have become important as essential tools for facilitating human and object interactions. However, the materials utilized for the electrodes of capacitive tactile sensors often cannot simultaneously exhibit high conductivity, low modulus, and strong adhesiveness. This limitation restricts their application on flexible interfaces and results in device failure due to mechanical mismatch. Herein, we report an ultra-low modulus, highly conductive, and adhesive elastomer and utilize it to fabricate a microstructure-coupled multifunctional flexible tactile sensor. We prepare a supramolecular conductive composite film (SCCF) as the electrode of the tactile sensor using a supramolecular deep eutectic solvent, polyvinyl alcohol (PVA) solution, poly(3,4-ethylenedioxythiophene):poly(styrene sulfonate) (PEDOT:PSS), and MXene suspension. We employ a polyvinylidene fluoride–hexafluoropropylene (PVDF–HFP) film containing 1-ethyl-3-methylimidazolium bis(trifluoromethylsulfonyl)imide (EMIM:TFSI) as the dielectric layer to fabricate capacitive sensors with an electrical double layer structure. Furthermore, we enhance the performance of the device by incorporating coupled pyramid and dome microstructures, which endow the sensor with multi-directional force detection. Our SCCF exhibits extremely high conductivity (reaching 710 S cm−1), ultra-low modulus (0.8 MPa), and excellent interface adhesion strength (>120 J m−2). Additionally, due to the outstanding conductivity and unique structure of the SCCF, it possesses remarkable electromagnetic shielding ability (>50 dB). Moreover, our device demonstrates a high sensitivity of up to 1756 kPa−1and a wide working range reaching 400 kPa, combining these attributes with the requirements of an ultra-soft human–machine interface to ensure optimal contact between the sensor and interface materials. This innovative and flexible tactile sensor holds great promise and potential for addressing various and complex demands of human–machine interaction.

Objective: Uterine corpus endometrial carcinoma (UCEC), a kind of gynecologic malignancy, poses a significant risk to women’s health. The precise mechanism underlying the development of UCEC remains elusive. Zinc finger protein 554 (ZNF554), a member of the Krüppel-associated box domain zinc finger protein superfamily, was reported to be dysregulated in various illnesses, including malignant tumors. This study aimed to examine the involvement of ZNF554 in the development of UCEC. Methods: The expression of ZNF554 in UCEC tissues and cell lines were examined by qRT-PCR and Western blot assay. Cells with stably overexpressed or knocked-down ZNF554 were established through lentivirus infection. CCK-8, wound healing, and Transwell invasion assays were employed to assess cell proliferation, migration, and invasion. Propidium iodide (PI) staining combined with fluorescence-activated cell sorting (FACS) flow cytometer was utilized to detect cell cycle distribution. qRT-PCR and Western blotting were conducted to examine relative mRNA and protein levels. Chromatin immunoprecipitation assay and luciferase reporter assay were used to explore the regulatory role of ZNF554 in RNA binding motif 5 (RBM5). Results: The expression of ZNF554 was found to be reduced in both UCEC samples and cell lines. Decreased expression of ZNF554 was associated with higher tumor stage, decreased overall survival, and reduced disease-free survival in UCEC. ZNF554 overexpression suppressed cell proliferation, migration, and invasion, while also inducing cell cycle arrest. In contrast, a decrease in ZNF554 expression resulted in the opposite effect. Mechanistically, ZNF554 transcriptionally regulated RBM5, leading to the deactivation of the Wingless (WNT)/β-catenin signaling pathway. Moreover, the findings from rescue studies demonstrated that the inhibition of RBM5 negated the impact of ZNF554 overexpression on β-catenin and p-glycogen synthase kinase-3β (p-GSK-3β). Similarly, the deliberate activation of RBM5 reduced the increase in β-catenin and p-GSK-3β caused by the suppression of ZNF554. In vitroexperiments showed that ZNF554 overexpression-induced decreases in cell proliferation and migration were counteracted by RBM5 knockdown. Additionally, when RBM5 was overexpressed, it hindered the improvements in cell proliferation and migration caused by reducing the ZNF554 levels. Conclusion: ZNF554 functions as a tumor suppressor in UCEC. Furthermore, ZNF554 regulates UCEC progression through the RBM5/WNT/β-catenin signaling pathway. ZNF554 shows a promise as both a prognostic biomarker and a therapeutic target for UCEC.