Multilevel Structure Research Articles

Detecting differential item functioning in presence of multilevel data: do methods accounting for multilevel data structure make a DIFference?

Assessment practices are, among other things, concerned with issues of fairness and appropriate score interpretation, in particular when making claims about subgroup differences in performance are of interest. In order to make such claims, a psychometric concept of measurement invariance or differential item functioning (DIF) ought to be considered and met. Over the last decades, researchers have proposed and developed a plethora of methods aimed at detecting DIF. However, DIF detection methods that allow multilevel data structures to be modeled are limited and understudied. In the current study, we evaluated the performance of four methods, including the model-based multilevel Wald and the score-based multilevel Mantel–Haenszel (MH), and two well-established single-level methods, the model-based single-level Lord and the score-based single-level MH. We conducted a simulation study that mimics real-world scenarios. Our results suggested that when data were generated as multilevel, mixed results regarding performances were observed, and not one method consistently outperformed the others. Single-level Lord and multilevel Wald yielded best control of the Type I error rates, in particular in conditions when latent means were generated as equal for the two groups. Power rates were low across all four methods in conditions with small number of between- and within-level units and when small DIF was modeled. However, in those conditions, single-level MH and multilevel MH yielded higher power rates than either single-level Lord or multilevel Wald. This suggests that current practices in detecting DIF should strongly consider adopting one of the more recent methods only in certain contexts as the tradeoff between power and complexity of the method may not warrant a blanket recommendation in favor of a single method. Limitations and future research directions are also discussed.

Factor Structure, Reliability, and Convergent Validity of an Ecological Momentary Assessment Binge-Eating Symptoms Scale.

Ecological momentary assessment (EMA) of binge-eating symptoms has deepened our understanding of eating disorders. However, there has been a lack of attention on the psychometrics of EMA binge-eating symptom measures. This paper focused on evaluating the psychometric properties of a four-item binge-eating symptom measure, including multilevel factor structure, reliability, and convergent validity. Forty-nine adults with binge-eating disorder and/or food addiction completed baseline questionnaires and a 10-day EMA protocol. During EMA, participants completed assessments of eating episodes, including four binge-eating symptom items. Analyses included multilevel exploratory factor analysis, computation of omega and intraclass correlation coefficients, and multilevel structural equation models of associations between contextual factors and binge-eating symptoms. A one within-subject factor solution fit the data and showed good multilevel reliability and adequate within-subjects variability. EMA binge-eating symptoms were associated with baseline binge-eating measures as well as relevant EMA eating characteristics: including greater unhealthful food and drink intake; higher perceived taste of food; lower likelihood to be planned eating; and lower likelihood of eating to occur at work/school and other locations and greater likelihood to occur at restaurants compared to home. In conclusion, the study findings support the psychometrics of a 4-item one-factor EMA measure of binge-eating symptoms.

Metal–organic framework–derived trimetallic particles encapsulated by ultrathin nitrogen-doped carbon nanosheets on a network of nitrogen-doped carbon nanotubes as bifunctional catalysts for rechargeable zinc–air batteries

Economical oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) bifunctional catalysts with high activity aimed at replacing precious metal catalysts for rechargeable zinc-air batteries (ZABs) must be developed. In this study, a multiple hierarchical-structural material is developed using a facile dielectric barrier discharge (DBD) plasma surface treatment, solvothermal reaction, and high-temperature carbonization strategy. This strategy allows for the construction of nanosheets using nitrogen-doped carbon (NC) material-encapsulated ternary CoNiFe alloy nanoparticles (NPs) on a network of NC nanotubes (NCNTs), denoted as CoNiFe–NC@p–NCNTs. Precisely, the presence of abundant CoNiFe alloy NPs and the formation of M–N–C active sites created by transition metals (cobalt, nickel, and iron) coupled with NC can provide superior OER/ORR bifunctional properties. Moreover, the prepared NC layers with a multilevel pore structure contribute to a larger specific surface area, exposing numerous active sites and enhancing the uniformity of electron and mass movement. The CoNiFe0.08–NC@p–NCNTs show remarkable dual functionality for electrochemical oxygen reactions (ORR half-wave potential of 0.811 V, limiting current density of 5.73 mA cm−2 measured with a rotating disk electrode at a rotation speed of 1600 rpm, and OER overpotential of 351 mV at 10 mA cm−2), which demonstrates similar ORR performance to 20 wt% Pt/C and better OER performance than the commercial RuO2. A liquid ZAB prepared using the proposed material has excellent bifunctionality with an open-circuit voltage of 1.450 V and long-term cycling stability of 230 h@10 mA cm−2.

Multistage structural self-supported electrode via structural engineering on carbon nanofibers for efficient hydrogen evolution reaction at high current density

Pt-based catalysts suffered from the wastage of active sites, resulting in unsatisfactory actual catalytic performance and poor durability as non-self-supporting catalysts for alkaline hydrogen evolution reaction (HER). In this work, we designed a self-supporting electrode material with multi-level structure that utilized carbon nanofibers anchored with cobalt to load and stabilize Pt nanoparticles (Pt@Co/CNFs) for achieving high activity and stability in alkaline HER. The carbon nanofibers served as a framework, with cobalt nanoparticles as an electron donor that can transfer electrons to the metal Pt to accelerating water splitting. In addition, the morphology of the multi-level structure surface was controlled by the content of the reducing agent to giving the material excellent superhydrophilic/underwater superaerophobic properties with ultra-low Pt load (0.29 mg cm−2), which was beneficial for HER. HER test results showed that Pt@Co/CNFs exhibited great catalytic activity, with η10 = 15 mV and η100 = 101 mV. Besides, the stable mechanical structure and the stability of the designed catalytic surface and electron donor in the multi-level structure gave the material exceptional stability. During the 100-hour stability test at 100 mA cm−2, the voltage change was only 50 mV. Besides, utilizing the Pt@Co/CNFs self-supporting electrode directly in the AEM water electrolysis system, a cell voltage as low as 2.18 V was achieved at 500 mA cm−2. Impressively, it demonstrated operational stability for over 200 h without noticeable decay. This work provides a promising approach for the development of advanced self-supporting electrode catalysts.

Re-Entrant Green Scheduling Problem of Bearing Production Shops Considering Job Reworking

To solve various reworking and repair problems caused by unqualified bearing product quality inspections, this paper introduces a green re-entrant scheduling optimization method for bearing production shops considering job reworking. By taking into account quality inspection constraints, this paper establishes an integrated scheduling mathematical model based on the entire processing–transportation–assembly process of bearing production shops with the goals for minimizing the makespan, total carbon emissions, and waste emissions. To solve these problems, the concepts of the set of the longest common machine routes (SLCMR) and the set of the shortest recombination machine combinations (SSRMC) were used to propose the re-entrant scheduling optimization method, based on system reconfiguration, to enhance the system stability and production scheduling efficiency. Then, a multi-objective hybrid optimization algorithm, based on a neighborhood local search (MOOA-LS), is proposed to improve the search scope and optimization ability by constructing a multi-level neighborhood search structure. Finally, this paper takes a bearing production shop as an example to carry out the case study and designs a series of experimental analyses and comparative tests. The final results show that in the bearing production process, the proposed model and algorithm can effectively realize green and energy-saving re-entrant manufacturing scheduling.

Zif-derived Co@hollow carbon nanofibers boost CO2 chemical fixation

In pursuit of carbon neutrality, the development of highly efficient catalysts for the CO2 cycloaddition with epoxides represents a prominent approach of utilizing CO2 to produce valuable cyclic carbonates. Herein, Co@hollow carbon nanofibers (Co@HCNFs) catalysts with multi-level pore structures were fabricated by coaxial electrospinning, in-situ growth of zeolitic imidazolate framework-67 (ZIF-67) and subsequent high temperature pyrolysis. The hollow structure and the formation of the active sites of Co@N-doped carbon (Co@CN) were regulated by changing the feeding ratio of core to shell and the Co(NO3)2·6H2O concentration during the ZIF-67 growth, thereby fine-tuning the catalyst structures and its catalytic performance. The as-prepared Co@HCNFs-0.6–0.1 catalyst exhibits superior catalytic performance in the CO2 cycloaddition and epichlorohydrin without a co-catalyst, achieving superior specific activity of 838 mmol⋅min−1⋅g−1. The outstanding catalytic performance is attributed to its multi-level pore structures with rich N content and well-distributed Co@CN nanoparticles. Moreover, Co@HCNFs-0.6–0.1 exhibits good stability over ten reaction cycles.

Preparation and moisture transport behavior of multilevel branching network nanofiber membrane: Based on the Lattice Boltzmann method and inspired by plant transpiration

Ensuring the efficiency and safety of firefighters relies heavily on optimizing the moisture permeability of their suits. Inspired by the intricate multi-level network structure of plant transpiration, this study utilized electrospinning technology to prepare biomimetic vascular, veined multilevel branched network quick-drying nanofiber membranes. Combining X-ray microtomography (XMT) and the lattice Boltzmann method (LBM), this study conducted a 3D reconstruction of the microstructure of quick-drying nanofiber materials, establishing a characterization method for moisture permeability behavior from a micro to mesoscale perspective and conducted numerical simulations. The results indicate that in the nanofiber membrane, the fiber diameter and pore size between fibers gradually decrease from the inlet to the outlet in the forward flow direction (from PET side to PAN side). At the same displacement pressure difference, the average and maximum flow rates of forward percolation were significantly greater than reverse percolation (from PAN side to PET side); in the initial stage of forward percolation (x < 11), the flow rate first increased and then gradually decreased, with the rate of increase (k = 0.712623) significantly higher than the rate of decrease (k'=0.001102), while in reverse percolation, the flow rate continuously increased throughout the process; These phenomena are primarily due to different startup pressure gradients. Furthermore, the analysis of moisture transport performance indicates that the fiber membrane exhibits highly efficient unidirectional moisture transport capability. This study expands the understanding of the moisture permeability mechanism of nanofibers, which is significant for enhancing the protective performance and comfort of quick-drying firefighter fabrics.

Hierarchical Model Rule Based Nlp for Semantic Training Representation Using Multi Level Structures

Hierarchical Model Rule Based Nlp for Semantic Training Representation Using Multi Level Structures

CodeTree: A System for Learnersourcing Subgoal Hierarchies in Code Examples

Subgoal-labeled code examples help learners understand code patterns and apply them to different problem contexts. Subgoal labels are multi-level in nature and based on goal structures that define the hierarchical functional units in code. Data-driven methods and experts can supply the goal structures, but they do not work in environments with scarce data and limited availability of experts. Previous research has shown that learnersourcing is effective for sourcing high-quality subgoal labels of given goal structures. We extend this research by learnersourcing goal structures themselves, thereby making the generation of subgoal-labeled materials fully learner-driven. We introduce CodeTree, a system that generates multi-level goal structures by aggregating learner-generated subgoals from two subgoal learning activities---Generation and Selection. In a between-subjects study, 45 novices studied three code examples with either CodeTree or code explanations alone. The results showed that CodeTree could learnersource high-quality goal structures and subgoal labels for all three examples with just five learners. Learners reported a significantly higher learning gain and satisfaction compared to the baseline.

Facile design of amphiphobic PVDF membranes utilizing synergy of pore structure and surface fluorination: A demonstration in membrane distillation

Amphiphobic membranes have attracted much attention in membrane distillation (MD) applications, due to its potential in simultaneously resisting membrane wetting and fouling when treating complex wastewater. The current amphiphobic membranes often require non-trivial design of multilevel re-entrant structure. Here we proposed a new class of amphiphobic membranes without multilevel re-entrant surface, by creating a composite poly(vinylidene fluoride)/polyvinyl alcohol/SiO2 membrane with asymmetric structure and well-controlled small surface pores via the NTIPS method, followed by direct fluorination via chemically binding the –OH of polyvinyl alcohol and low surface energy moieties 1H,1H,2H,2H-perfluorodecyl-triethoxysilane, assisted by 3-aminopropyltriethoxysilane and trimesoyl chloride. The amphiphobic membrane showed superior liquid entry pressure above 5 bar with various simulated saline feeds containing surfactants and organic pollutants. The amphiphobic membrane demonstrated robust anti-wetting and anti-fouling performance through a series of 24-h direct contact MD experiments using the above feeds, presenting stable water flux and high rejection, e.g., flux of 50.6 kg/(m2·h) and rejection >99.9 % with feed containing 3.5 wt% NaCl, 1 mM FA, and 2 mM SDS. The results demonstrated the fabrication of robust amphiphobic membrane through simple procedure utilizing the synergy between pore structure and surface chemistry, inspiring more efforts to design adequate membranes for treating complex wastewater.

Conditioning on the Pre-Test versus Gain Score Modelling: Revisiting the Controversy in a Multilevel Setting.

We consider estimating the effect of a treatment on a given outcome measured on subjects tested both before and after treatment assignment in observational studies. A vast literature compares the competing approaches of modelling the post-test score conditionally on the pre-test score versus modelling the difference, namely, the gain score. Our contribution lies in analyzing the merits and drawbacks of two approaches in a multilevel setting. This is relevant in many fields, such as education, where students are nested within schools. The multilevel structure raises peculiar issues related to contextual effects and the distinction between individual-level and cluster-level treatments. We compare the two approaches through a simulation study. For individual-level treatments, our findings align with existing literature. However, for cluster-level treatments, the scenario is more complex, as the cluster mean of the pre-test score plays a key role. Its reliability crucially depends on the cluster size, leading to potentially unsatisfactory estimators with small clusters.

Testing similarity in longitudinal networks: The Individual Network Invariance Test.

The comparison of idiographic network structures to determine the presence of heterogeneity is a challenging endeavor in many applied settings. Previously, researchers eyeballed idiographic networks, computed correlations, and used techniques that make use of the multilevel structure of the data (e.g., group iterative multiple model estimation and multilevel vector autoregressive) to investigate individual differences. However, these methods do not allow for testing the (in)equality of idiographic network structures directly. In this article, we propose the Individual Network Invariance Test (INIT), which we implemented in the R package INIT. INIT extends common model comparison practices in structural equation modeling to idiographic network structures to test for (in)equality between idiographic networks. In a simulation study, we evaluated the performance of INIT on both saturated and pruned idiographic network structures by inspecting the rejection rate of the χ² difference test and model selection criteria, such as the Akaike information criterion (AIC) and Bayesian information criterion (BIC). Results show INIT performs adequately when t = 100 per individual. When applying INIT on saturated networks, the AIC performed best as a model selection criterion, while the BIC showed better results when applying INIT on pruned networks. In an empirical example, we highlight the possibilities of this new technique, illustrating how INIT provides researchers with a means of testing for (in)equality between idiographic network structures and within idiographic network structures over time. To conclude, recommendations for empirical researchers are provided. (PsycInfo Database Record (c) 2024 APA, all rights reserved).

UCDCN: a nested architecture based on central difference convolution for face anti-spoofing

The significance of facial anti-spoofing algorithms in enhancing the security of facial recognition systems cannot be overstated. Current approaches aim to compensate for the model’s shortcomings in capturing spatial information by leveraging spatio-temporal information from multiple frames. However, the additional branches to extract inter-frame details increases the model’s parameter count and computational workload, leading to a decrease in inference efficiency. To address this, we have developed a robust and easily deployable facial anti-spoofing algorithm. In this paper, we propose Central Difference Convolution UNet++ (UCDCN), which takes advantage of central difference convolution and improves the characterization ability of invariant details in diverse environments. Particularly, we leverage domain knowledge from image segmentation and propose a multi-level feature fusion network structure to enhance the model’s ability to capture semantic information which is beneficial for face anti-spoofing tasks. In this manner, UCDCN greatly reduces the number of model parameters as well as achieves satisfactory metrics on three popular benchmarks, i.e., Replay-Attack, Oulu-NPU and SiW.

Foreign Experience Review on Engineering Protection of Seashores and Hillslopes

Introduction. During design, construction and operation of buildings and structures in the coastal areas there arises a serious problem of engineering protection of coasts and hillslopes from the dangerous geological processes, which include the coastal erosion, underflooding of territories, hillslope erosion, gravitational (slope) processes, such as intensification of landslides, etc. These processes often result in significant economic damage usually related to the non-recoverable loss of valuable coastal territories, deformation of buildings and structures, damage and destruction of the objects of transport and engineering infrastructure. In this regard, the study of the advanced foreign experience in engineering protection of the coastal areas and the evaluation of the possibility of it to be implemented in Russia is a relevant scientific and engineering task.Materials and methods. To prepare the review, the following data was used: field observation data with photographic evidences of the objects of engineering protection of seashores and hillslopes obtained by the authors during a business trip to the People's Republic of China (PRC) in October–November 2023; study and analysis of literature sources in the subject area, including the methods and technologies implemented in the Netherlands, Japan, the USA, Great Britain, Italy; summary and systematisation of the coastal protection methods for further development of their classification and evaluation of the possibility of implementing thereof in the Russian Federation.Research results. The main principles of design, the advanced methods and technologies of coastal protection used in China have been defined. They include: construction of the multi-level protective structures, such as the multiple-row breakwaters of special shape in combination with the stepped retaining walls, the vertical walls consisting of the driven piles, anchoring and spray-concrete covering of the slopes in combination with the metal meshes; terracing in combination with the retaining reinforced concrete (horizontal and vertical) crossbeams; the sloping wave-absorbing reinforced concrete surfaces; the meshy structures and fences combating the Aeolian processes, etc.The main methods used in the Netherlands, Japan, the USA and Italy for comprehensive protection of the large territories have been defined. Such as: the network of dams, weirs, wave breakers, coast paving, regulation of the flows by water locks and barriers, creation of the artificial protective islands; artificial replenishment with sand; creation of the green spaces along the coast; forming the dunes — natural or artificial hills of sand or pebbles located along the coast; creation of the concrete structures, stone jetties, floating constructions and even artificial reefs.The importance of management was emphasised, including the early warning systems for storms and tsunamis, weather and marine condition monitoring systems, as well as dissemination of the warning messages and evacuation plans among the population in the areas of potential threat.Discussion and conclusion. Based on the results of the review, the main conclusions were formulated. The recommendations were given on the possible ways of improving the coastal protection of the Caucasian Black Sea Coast and other coastal areas of the Russian Federation based on the advanced foreign experience.

M3W: Multistep Three-Way Clustering.

Three-way clustering has been an active research topic in the field of cluster analysis in recent years. Some efforts are focused on the technique due to its feasibility and rationality. We observe, however, that the existing three-way clustering algorithms struggle to obtain more information and limit the fault tolerance excessively. Moreover, although the one-step three-way allocation based on a pair of fixed, global thresholds is the most straightforward way to generate the three-way cluster representations, the clusters derived from a pair of global thresholds cannot exactly reveal the inherent clustering structure of the dataset, and the threshold values are often difficult to determine beforehand. Inspired by sequential three-way decisions, we propose an algorithm, called multistep three-way clustering (M3W), to address these issues. Specifically, we first use a progressive erosion strategy to construct a multilevel structure of data, so that lower levels (or external layers) can gather more available information from higher levels (or internal layers). Then, we further propose a multistep three-way allocation strategy, which sufficiently considers the neighborhood information of every eroded instance. We use the allocation strategy in combination with the multilevel structure to ensure that more information is gradually obtained to increase the probability of being assigned correctly, capturing adaptively the inherent clustering structure of the dataset. The proposed algorithm is compared with eight competitors using 18 benchmark datasets. Experimental results show that M3W achieves superior performance, verifying its advantages and effectiveness.

Direct adjusted comparison of expressed emotion towards patients with schizophrenia between halfway houses and family settings

IntroductionRates of high expressed emotion (EE) towards patients with schizophrenia have only indirectly been compared between families and community residential facilities, since studies including patients in both settings are unfortunately lacking. High EE rates in staff-patient studies are typically lower than in families, with negligible rates of high emotional overinvolvement (EOI). However, indirect comparisons can suffer from many biases.ObjectivesThis study directly compared patients with schizophrenia living in halfway houses or with their families on the EE of their caregivers, adjusting for patient- and caregiver-related confounders.MethodsWe included 40 inpatients with schizophrenia living in halfway houses and 40 outpatients living with their families and recorded the EE of the caring staff (N=22 nurses) or parents (N=56), respectively, through Five Minutes Speech Sample interviews. Each nurse rated 1-12 inpatients and each inpatient was rated by 2-5 nurses, totaling 155 nurse ratings. Each outpatient was rated by one or both parents. Due to the multilevel structure of EE ratings, generalized linear mixed models were fitted. We first adjusted only for differences in patient-related confounders between groups and then added basic caregiver-related demographics.ResultsCompared to outpatients, inpatients were older (p=0.001), less well educated (p=0.002), had a longer disease duration (p=0.047), more hospitalizations (p=0.012), lower severity of psychotic (p=0.027) and, specifically, negative symptoms (p=0.015), and lower perceived criticism (p=0.001). Nurses were younger (p&lt;0.001) and better educated (p=0.001) than parents. After adjusting for patient-related confounders only, EOI was significantly higher in parents (p=0.027) while criticism did not significantly differ between groups. However, after also adjusting for caregiver demographics (age, gender and education), criticism was significantly higher in nurses (p=0.027) while differences in EOI became non-significant.ConclusionsDifferences in EE, when directly compared between parents and professional caregivers, may be explained by differences in patient-related characteristics, caregiver demographics as well as other caregiver characteristics to be investigated in future studies.Disclosure of InterestNone Declared

The Layout of the Ecological Circle of “Insurance &amp; Pension Service” in China in the Longevity Era

With the deepening of aging, profound changes are taking place in the population structure. The demand structure of the contemporary elderly is changing from survival to development, paying more attention to the quality of life and spiritual needs, and the pension cycle of the elderly is long-term, dynamic, including physical and psychological. In recent years, China has continued to promote the construction of a multi-level and multi-pillar old-age security system. Old-age finance, as one of the “five major articles”, has become the key work of the financial industry after it was proposed at the central financial work conference. How does the insurance industry play an important role in the development of old-age finance, build the ecological pattern and system of high-quality old-age services under the background of silver economy, and accurately innovate and build a multi-level structure of old-age security guided by the needs of the elderly? It has become an urgent content to be solved. Starting from the current situation of China's pension service system, the demand-side study found that China is facing an increasingly serious problem of population aging, and the supply side grasps the “first year” of high-quality construction of the old-age service system. This paper puts forward some countermeasures and suggestions on the layout of the ecological circle of “insurance &amp; pension service” in China.

Russian experience in managing interregional cooperation

The article analyzes Russia's experience in managing interregional cooperation, focusing on key concepts and regulatory support, as well as management bodies involved in coordinating interactions between regions. Legislative acts regulating the processes of cooperation and the activities of both state and local management agencies are considered, which allows a deeper assessment of their role in the implementation of interregional projects. Special attention is paid to successful examples such as Ural-Siberia and North-West, which contribute to the development of transport corridors and cultural ties. The article reveals the functions of government bodies, emphasizing the multilevel structure of management of interregional interactions in Russia, which is critical for economic growth, social integration and strengthening the unity of the country. The methodology allows to identify the key factors influencing the effectiveness of interregional cooperation, as well as to assess the role of management structures in the implementation of joint projects. The scientific novelty of the article lies in the systematic analysis of interregional cooperation in Russia, which allows us to identify its unique characteristics and trends. Unlike existing studies, this work focuses on the interaction of federal and local authorities, as well as on the analysis of legislation regulating interregional relations. The findings of the study emphasize that effective interregional cooperation requires an integrated approach, including improved coordination between management levels, the development of a unified legal model and the active participation of regions in federal initiatives. The article also reveals the importance of financing interregional programs, demonstrating their positive impact on the socio-economic development of regions, which is confirmed by concrete examples of successful projects and agreements. Thus, the work makes a significant contribution to understanding the mechanisms and dynamics of interregional cooperation.

MGCoT: Multi-Grained Contextual Transformer for table-based text generation

Recent advances in Transformer have led to the revolution of table-based text generation. However, most existing Transformer-based architectures ignore the rich contexts among input tokens distributed in multi-level units (e.g., cell, row, or column), leading to sometimes unfaithful text generation that fails to establish accurate association relationships and misses vital information. In this paper, we propose Multi-Grained Contextual Transformer (MGCoT), a novel architecture that fully capitalizes on the multi-grained contexts among input tokens and thus strengthens the capacity of table-based text generation. The key primitive, Multi-Grained Contexts (MGCo) module, involves two components: a local context sub-module that adaptively gathers neighboring tokens to form the token-wise local context features, and a global context sub-module that consistently aggregates tokens from a broader range to form the shared global context feature. The former aims at modeling the short-range dependencies that reflect the salience of tokens within similar fine-grained units (e.g., cell and row) attending to the query token, while the latter aims at capturing the long-range dependencies that reflect the significance of each token within similar coarse-grained units (e.g., multiple rows or columns). Based on the fused multi-grained contexts, MGCoT can flexibly and holistically model the content of a table across multi-level structures. On three benchmark datasets, ToTTo, FeTaQA, and Tablesum, MGCoT outperforms strong baselines by a large margin on the quality of the generated texts, demonstrating the effectiveness of multi-grained context modeling. Our source codes are available at https://github.com/Cedric-Mo/MGCoT.

Construction of multi-porous Ni-based monolithic catalyst by combining wood carbon and SBA-15 for strengthening CO2 methanation

Monolithic catalysts meet the needs of practical application compared with powder catalysts. In this work, we constructed a Ni-based monolithic catalyst (Ni/SAB-15/WWC) by in situ growth of rod-like mesoporous SBA-15 in the microchannels of wood carbon. The small Ni-based nanoparticles (∼5.3 nm) were uniformly dispersed into the mesoporous channels via a post-loading method. The monolithic Ni/SAB-15/WWC catalyst with a multilevel porous structure could provide a laminar gas flow on the surface of Ni active sites, increasing the low-temperature activity and high-temperature durability of CO2 methanation. The Ni/SAB-15/WWC with low Ni loading (∼2%) displayed superior CO2 methanation catalytic performance in terms of low-temperature activity and CH4 selectivity compared with the powder Ni/SAB-15 with high Ni loading (∼15 %). The combination of wood carbon and mesoporous materials is an innovative strategy for the design monolithic catalysts with low metal loading and excellent catalytic performance.