This study examines the reliability and validity of AI-generated scoring for continuation writing tasks.By comparing GPT-4 with eight experienced human raters across 21 student responses,it evaluates AI’s consistency...This study examines the reliability and validity of AI-generated scoring for continuation writing tasks.By comparing GPT-4 with eight experienced human raters across 21 student responses,it evaluates AI’s consistency,severity,and alignment with human scoring criteria.Results show that AI exhibits high self-consistency and adapts effectively to different scoring roles(e.g.,teacher vs.highstakes rater).However,AI scores were more lenient than human raters and demonstrated divergent evaluation focuses—prioritizing narrative coherence and emotional depth,while teachers emphasized linguistic accuracy and richness of detail.The findings suggest AI’s potential as a supplementary assessment tool,offering rapid,holistic feedback,but highlight the need for further calibration to align with educational standards.Implications include exploring hybrid evaluation models that leverage the strengths of both AI and human raters to achieve more equitable,efficient,and pedagogically meaningful writing assessments.展开更多
This study investigated the effects of xu-argument-based continuation writing on learners’processing of source texts.Seventy-five participants were randomly assigned to three conditions:(1)continuation writing,(2)sum...This study investigated the effects of xu-argument-based continuation writing on learners’processing of source texts.Seventy-five participants were randomly assigned to three conditions:(1)continuation writing,(2)summary writing,or(3)reading comprehension.Eye-tracking data were collected during reading,measuring early(first fixation duration,first pass duration)and late(go-past time,total fixation duration)eye movements.During writing,source-text rereading was tracked via fixation counts and durations.Results showed that task type did not affect initial lexical access,as first fixation duration showed no group differences.However,both production groups exhibited significantly longer first pass durations than the reading comprehension group.Late measures revealed a gradient pattern:the continuation writing group spent significantly longer gopast time and total fixation duration than the summary writing group,which exceeded the reading comprehension group.This indicates that continuation tasks promoted deeper cognitive engagement during reading.During writing,the continuation writing group spent more time rereading the source text with higher fixation counts than the summary writing group.These findings suggest that continuation writing triggers more intensive reader-text interaction during pre-writing and enhances comprehension-production coupling through sustained attention to input during writing.This study sheds light on the cognitive mechanisms underlying the theoretical and pedagogical value of xu-argument.展开更多
BACKGROUND:This study is to evaluate clearance eff ects of hemoperfusion(HP),continuous renal replacement therapy(CRRT),and plasma exchange(PE)for chlorfenapyr and its metabolite tralopyril in patients with acute pois...BACKGROUND:This study is to evaluate clearance eff ects of hemoperfusion(HP),continuous renal replacement therapy(CRRT),and plasma exchange(PE)for chlorfenapyr and its metabolite tralopyril in patients with acute poisoning.METHODS:This retrospective study included 18 patients with acute oral chlorfenapyr poisoning treated at our department between January 2022 and January 2024.All patients received conventional therapies combined with blood purification,including HP,CRRT,and PE.HP was performed three sessions within the fi rst 24 h,followed by CRRT and PE.Serial blood samples were collected to measure plasma concentrations of chlorfenapyr and tralopyril using gas chromatography/liquid chromatography-mass spectrometry(GC/LC-MS).The toxin-clearance eff ects were assessed using a linear mixed-eff ects(LME)model.RESULTS:The hourly decline rate of the plasma chlorfenapyr concentration(median[IQR])was 8.83%(1.79%)for HP,4.12%(1.26%)for CRRT,and 6.85%(1.44%)for PE.LME analysis showed higher decline rate in the plasma concentration with HP(β=5.00;P<0.001)and PE(β=2.15;P=0.003)compared to CRRT.For tralopyril,the hourly decline rates were 3.04%(0.62%)for HP,1.82%(0.48%)for CRRT,and 3.01%(0.37%)for PE.LME analysis showed that the clearance effects of HP(β=0.027;P<0.001)and PE(β=0.022;P=0.001)were superior to CRRT.Pre-treatment toxin levels and the interval from hospital admission to blood purifi cation showed no signifi cant interaction with clearance outcomes.CONCLUSION:In our study,HP was associated with a higher decline rate in plasma chlorfenapyr concentration compared to CRRT and PE,supporting HP as a preferred early intervention.However,all three methods showed limited effi cacy in reducing tralopyril levels.Further research into the toxicokinetics and mechanisms of chlorfenapyr is warranted to optimize purifi cation strategies.展开更多
Background:Continuous care for children with enterostomy and their families has been gaining popularity in China.Objective:To evaluate the feasibility of continuous care for children with enterostomy and their familie...Background:Continuous care for children with enterostomy and their families has been gaining popularity in China.Objective:To evaluate the feasibility of continuous care for children with enterostomy and their families in China.Methods:The PubMed,Web of Science,Embase,Cochrane Library,EBSCO,CNKI,CBM,VIP,and WanFang were searched for clinical trials until December 30,2025.Two reviewers independently searched articles,evaluated quality and extracted data.This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses(PRISMA).Results:33 studies involving 2774 participants were included.The meta-analysis showed that continuous care strategy can significantly reduce the incidence of complications in children with enterostomy(OR=0.20,95%CI=0.16-0.26,p<0.001,I2=0%),effectively improve the family caregiver ability for enterostomy(MD=-10.34,95%CI=-13.82 to-6.85,p<0.001,I2=99%),shorten the time for family members to replace stoma bags(MD=-13.57,95%CI=-19.66 to-7.49,p<0.001,I2=100%),and alleviate negative emotions such as anxiety(SMD=-1.80,95%CI=-2.36 to-1.23,p<0.001,I2=92%)and depression(SMD=-1.54,95%CI=-2.04 to-1.04,p<0.001,I2=89%)in the families of the affected children.Conclusions:Continuous care can reduce complications of enterostomy in children,improve the family caregiver ability for enterostomy and alleviate negative emotions of family members such as anxiety and depression.展开更多
Local resonant acoustic metamaterials have broad applications in sound insulation,yet their single-configuration designs often exhibit limited and discontinuous bandgap widths,hindering full-frequency noise attenuatio...Local resonant acoustic metamaterials have broad applications in sound insulation,yet their single-configuration designs often exhibit limited and discontinuous bandgap widths,hindering full-frequency noise attenuation across the human auditory range.This study presents a double-phase fidget-spinner-shaped acoustic metamaterial(DFAM),specifically designed to achieve an ultra-broad,low-frequency continuous bandgap by means of synergistic structural optimization,enabling effective and robust control of audible noise.Based on Bloch's theorem and the finite element method,the dispersion relation of the DFAM structure is calculated and verified by the transmission loss curves.The propagation characteristics of sound waves within the structure are further analyzed for noise frequencies that fall within the passband.The influence of the geometric and physical parameters on the bandgap is investigated,and the corresponding transmission loss in the propagation direction is further calculated.A hybrid collaborative design strategy,leveraging multi-parameter optimization and bandgap complementarity,is developed to construct a metastructure with continuous bandgap coverage from 20 Hz to 1000 Hz.The resulting metastructure demonstrates exceptional broadband noise attenuation,achieving a total bandgap width of 876.3 Hz(87.63% of the target range)with the transmission loss up to-762.78 d B in a three-periodic arrangement.The simulation and experimental results for the transmission loss of the DFAM metastructure show strong agreement in the low-frequency range.This work provides a novel framework for designing ultra-wide low-frequency continuous bandgap metastructures,offering significant potential for noise mitigation in complex environments.展开更多
The refinement of the as-cast grain structure in austenitic heat-resistant stainless steel depends on the formation of active solid nuclei during solidification.Titanium(Ti)additions successfully induced the formation...The refinement of the as-cast grain structure in austenitic heat-resistant stainless steel depends on the formation of active solid nuclei during solidification.Titanium(Ti)additions successfully induced the formation of Ti-containing inclusions,enhancing heterogeneous nucleation and promoting equiaxed dendritic growth in 347H stainless steel.Thermal simulation experiments indicated that the equiaxed crystal ratios increased notably with Ti content;samples with 0.06,0.12,and 0.36 wt.%Ti exhibited equiaxed ratios of 18%,24%,and 41%,respectively.Three primary inclusion types—TiN,Al_(2)O_(3)-TiN,and TiO_(x)-TiN—were identified at the cores of equiaxed dendrites,with nucleation core sizes predominantly ranging from 0.5 to 8μm.Among the tested samples,the 0.36 wt.%Ti addition produced the highest nucleation core density.Increasing Ti content significantly elevated dendrite tip undercooling from 2.6 K(0.06 wt.%Ti)to 10.8 K(0.36 wt.%Ti),accelerating solidification front instability and thus enhancing heterogeneous nucleation.Additionally,higher Ti content increased the divergence angle between adjacent columnar dendrites,further promoting the columnar-to-equiaxed transition(CET).展开更多
Introduction:The choice is no longer whether—but when and how!Today,organizations can no longer choose whether or not to engage with stakeholders;the only real decision is when and how to do it successfully.Engaging ...Introduction:The choice is no longer whether—but when and how!Today,organizations can no longer choose whether or not to engage with stakeholders;the only real decision is when and how to do it successfully.Engaging stakeholders is essential for any type of organization,public,private,or civil society.In fact,it is a fundamental part of effective governance,continuous improvement,and social responsibility.The principle behind stakeholder engagement is simple yet powerful:those who can influence or be affected by an organization's mission must be given the opportunity to express their views and contribute to shaping the decisions that impact them.展开更多
Continual learning fault diagnosis(CLFD)has gained growing interest in mechanical systems for its ability to accumulate and transfer knowledge in dynamic fault diagnosis scenarios.However,existing CLFD methods typical...Continual learning fault diagnosis(CLFD)has gained growing interest in mechanical systems for its ability to accumulate and transfer knowledge in dynamic fault diagnosis scenarios.However,existing CLFD methods typically assume balanced task distributions,neglecting the long-tailed nature of real-world fault occurrences,where certain faults dominate while others are rare.Due to the long-tailed distribution among different me-chanical conditions,excessive attention has been focused on the dominant type,leading to performance de-gradation in rarer types.In this paper,decoupling incremental classifier and representation learning(DICRL)is proposed to address the dual challenges of catastrophic forgetting introduced by incremental tasks and the bias in long-tailed CLFD(LT-CLFD).The core innovation lies in the structural decoupling of incremental classifier learning and representation learning.An instance-balanced sampling strategy is employed to learn more dis-criminative deep representations from the exemplars selected by the herding algorithm and new data.Then,the previous classifiers are frozen to prevent damage to representation learning during backward propagation.Cosine normalization classifier with learnable weight scaling is trained using a class-balanced sampling strategy to enhance classification accuracy.Experimental results demonstrate that DICRL outperforms existing continual learning methods across multiple benchmarks,demonstrating superior performance and robustness in both LT-CLFD and conventional CLFD.DICRL effectively tackles both catastrophic forgetting and long-tailed distribution in CLFD,enabling more reliable fault diagnosis in industrial applications.展开更多
We report an immobilized enzyme-catalyzed batch and continuous-flow synthesis of optically pure ethyl(R)-pantothenate((R)-PaOEt),the direct precursor of d-pantothenic acid.Firstly,a ketoreductase mutant designated as ...We report an immobilized enzyme-catalyzed batch and continuous-flow synthesis of optically pure ethyl(R)-pantothenate((R)-PaOEt),the direct precursor of d-pantothenic acid.Firstly,a ketoreductase mutant designated as M2,carrying two-point mutations of F97L and M242F relative to the wild-type SSCR,was constructed by site-directed mutagenesis,exhibited simultaneously improved activity toward ethyl 2′-ketopantothenate(K-PaOEt)and isopropanol,and could effectively catalyze the stereoselective reduction of K-PaOEt to(R)-PaOEt by using isopropanol as the sacrificial co-substrate to regenerate NADPH.After screening six commercially available carriers,an amino resin LXTE-700 was identified as the best solid support for the immobilization of M2 via the glutaraldehyde activation method.Upon optimization of the immobilization process and reaction conditions,the fabricated immobilized enzyme M2@amino resin demonstrated excellent recyclability and reusability,with the complete conversion of K-PaOEt to(R)-PaOEt being still realized after 12 cycles of reuse.Finally,M2@amino resin-catalyzed synthesis of(R)-PaOEt was successfully implemented in continuous-flow,accomplishing a 6.3 times higher space-time yield than that with the batch synthesis(529.2 versus 84 g L^(-1) d^(-1)).Our developed flow biocatalysis system also features an outstanding operational stability,as evidenced by the 100%conversion rate achieved after 15 consecutive days of operation.展开更多
1.Introduction.Pancreaticoduodenectomy is an essential surgical procedure for the treatment of malignant tumors in the pancreatic head,distal common bile duct,and duodenal papilla,and is widely used in clinical practi...1.Introduction.Pancreaticoduodenectomy is an essential surgical procedure for the treatment of malignant tumors in the pancreatic head,distal common bile duct,and duodenal papilla,and is widely used in clinical practice.One of the primary determinants of surgical success is the durability of the pancreaticojejunal anastomosis.展开更多
The iterative continuation task(ICT)requires English as a foreign language(EFL)learners to read a segment and write a continuation that aligns with the preceding segment of an English novel with successive turns,offer...The iterative continuation task(ICT)requires English as a foreign language(EFL)learners to read a segment and write a continuation that aligns with the preceding segment of an English novel with successive turns,offering exposure to diverse grammatical structures and opportunities for contextualized usage.Given the importance of integrating technology into second language(L2)writing and the critical role that grammar plays in L2 writing development,automated written corrective feedback provided by Grammarly has gained significant attention.This study investigates the impact of Grammarly on grammar learning strategies,grammar grit,and grammar competence among EFL college students engaged in ICT.This study employed a mixed-methods sequential exploratory design;56 participants were divided into an experimental group(n=28),receiving Grammarly feedback for ICT,and a control group(n=28),completing ICT without Grammarly feedback.Quantitative results revealed that both groups showed improvements in L2 grammar learning strategies,grit and competence.For the experimental group,significant differences were observed across all variables of L2 grammar learning strategies,grit,and competence between pre-and post-tests.For the control group,significant differences were only observed in the affective dimension of grammar learning strategies,Consistency of Interest(COI)of grammar grit,and grammar competence.However,the control group presented a significantly higher improvement in grammar competence.Qualitative analysis showed both positive and negative perceptions of Grammarly.The pedagogical implications of integrating Grammarly and ICT for L2 grammar development are discussed.展开更多
The transient phenomena of re-oxidation and slag entrapment occurring in the tundish during the ladle change-over process have been proven detrimental to clean steel production.Therefore,an unsteady three-phase turbul...The transient phenomena of re-oxidation and slag entrapment occurring in the tundish during the ladle change-over process have been proven detrimental to clean steel production.Therefore,an unsteady three-phase turbulence model,coupling velocity,temperature,and phase field was established to study the effect of the ladle shroud immersion depth on the slag eye formation,slag entrainment,slag dragging,air dragging,and flow characteristics during the ladle change-over process of a two-strand tundish.The results showed that reducing the immersion depth decreases the high-velocity region area under the slag layer in the quasi-steady process.During the emptying stage,as the molten bath level gradually decreases,the outlet temperature exhibits a trend of initially decreasing and subsequently increasing across all three shroud immersion depths.However,under a 210 mm shroud immersion depth,molten slag and air are dragged into the shroud,forming slag droplets and causing significant fluctuations,with a maximum scalar velocity of 0.0764 m/s at the monitoring point.In the filling stage,air and molten slag are dragged into the molten bath,forming bubbles and slag droplets at an immersion depth of 210 mm.Bubbles are observed within the molten slag layer,which can readily cause an emulsification phenomenon,making it easier to be dragged as slag droplets.Additionally,the slag eye area measured under 210 mm immersion depth at 45 s is 0.303 m^(2),while the maximum scalar velocity of 2.4259 m/s is detected at 12 s.At an immersion depth of 360 mm,the average area of the slag eye is minimized to 0.06268 m2,with corresponding variances of 0.006753,representing the optimal immersion depth.展开更多
A full-sectional microstructure characterization method was developed to investigate the formation of coarse slag rims during the continuous casting of hypo-peritectic steel.The cross-sectional microstructural analysi...A full-sectional microstructure characterization method was developed to investigate the formation of coarse slag rims during the continuous casting of hypo-peritectic steel.The cross-sectional microstructural analysis of typical slag rims for two highly crystalline powders revealed that their formation was primarily driven by the solidification of the liquid slag.Distinct differences were observed in the microstructures of slag rims from the two powders.Powder A(characterized by a higher breaking temperature and viscosity)displayed alternating lamellar microstructures of coarse and fine phases,with the coarse phases composed of akermanite-gehlenite transition phases.In contrast,powder B(with a lower breaking temperature and viscosity)predominantly comprised regular akermanite-gehlenite crystals interspersed with a certain amount of glassy phases.Numerical simulations of a three-phase fluid flow coupled with heat transfer indicate that slag rim formation correlates with mold oscillation.Solidification of the liquid slag at the slag rim front predominantly occurs during the negative stroke of the mold oscillation.The average heating rate during the ascending stage of the mold reaches approximately 100 K·s^(−1),whereas the average cooling rate during the descending stage attains 400 K·s^(−1).This temperature variation leads to the formation of lamellar microstructures,whereas the ascending stage promotes the formation of coarse structures and thicker slag rims.Based on the powder properties,two distinct formation pathways exist for highly crystalline mold powders.For the powders with a higher breaking temperature,higher viscosity,and narrower solidification range(powder A),coarse microstructures and thicker slag rims were preferentially formed.For powders with lower breaking temperature and viscosity and wider solidification ranges(powder B),the liquid slag resisted rapid solidification,and the extended mushy zone allowed the partial liquid slag to persist at the slag rim front,promoting the formation of a thin slag rim.This study enhances the understanding of slag rim formation in highly crystalline mold powders and provides critical insights into the control of longitudinal surface cracks in hypo-peritectic steel.展开更多
This study integrates explicit input enhancement into comparative continuation writing,defined as a task in which learners produce a continuation by comparing their own expression with an input text,aligning with its ...This study integrates explicit input enhancement into comparative continuation writing,defined as a task in which learners produce a continuation by comparing their own expression with an input text,aligning with its discourse structure and linguistic features,while developing their own ideas.It aims to examine whether English as a Foreign Language(EFL)learners in China exhibit differences in discourse competence and writing performance when completing comparative continuation writing combined with different input enhancement techniques,and whether the alignment effect occurs at the discourse level.Sixty first-year Chinese senior middle school students were divided into four groups:three groups engaged in comparative continuation writing with varying input enhancement,achieved by combining different techniques,while a control group performed a designated-topic writing task.The results revealed that three comparative continuation writing groups outperformed the designated-topic writing group in discourse competence,particularly in the use of temporal connectives.However,differences and some inconsistencies were observed among the comparative continuation writing groups across individual indices.The study highlights effective ways to incorporate comparative continuation writing into English instruction and demonstrates how explicit input enhancement can complement the task,simultaneously activating the alignment effect proposed by the xu-argument and enhancing discourse competence in writing.展开更多
Mental-health risk detection seeks early signs of distress from social media posts and clinical transcripts to enable timely intervention before crises.When such risks go undetected,consequences can escalate to self-h...Mental-health risk detection seeks early signs of distress from social media posts and clinical transcripts to enable timely intervention before crises.When such risks go undetected,consequences can escalate to self-harm,long-term disability,reduced productivity,and significant societal and economic burden.Despite recent advances,detecting risk from online text remains challenging due to heterogeneous language,evolving semantics,and the sequential emergence of new datasets.Effective solutions must encode clinically meaningful cues,reason about causal relations,and adapt to new domains without forgetting prior knowledge.To address these challenges,this paper presents a Continual Neuro-Symbolic Graph Learning(CNSGL)framework that unifies symbolic reasoning,causal inference,and continual learning within a single architecture.Each post is represented as a symbolic graph linking clinically relevant tags to textual content,enriched with causal edges derived from directional Point-wise Mutual Information(PMI).A two-layer Graph Convolutional Network(GCN)encodes these graphs,and a Transformer-based attention pooler aggregates node embeddings while providing interpretable tag-level importances.Continual adaptation across datasets is achieved through the Multi-Head Freeze(MH-Freeze)strategy,which freezes a shared encoder and incrementally trains lightweight task-specific heads(small classifiers attached to the shared embedding).Experimental evaluations across six diverse mental-health datasets ranging from Reddit discourse to clinical interviews,demonstrate that MH-Freeze consistently outperforms existing continual-learning baselines in both discriminative accuracy and calibration reliability.Across six datasets,MH-Freeze achieves up to 0.925 accuracy and 0.923 F1-Score,with AUPRC≥0.934 and AUROC≥0.942,consistently surpassing all continual-learning baselines.The results confirm the framework’s ability to preserve prior knowledge,adapt to domain shifts,and maintain causal interpretability,establishing CNSGL as a promising step toward robust,explainable,and lifelong mental-health risk assessment.展开更多
The paper considers the initial value problem of inhomogeneous fourth-order Schr¨odinger equation with potential in energy space H^(2)(R^(d)).The global well-posedness is obtained in dimensions d≥5 resorting to ...The paper considers the initial value problem of inhomogeneous fourth-order Schr¨odinger equation with potential in energy space H^(2)(R^(d)).The global well-posedness is obtained in dimensions d≥5 resorting to contractive mapping principle,Strichartz estimates,Caffarelli-Kohn-Nirenberg-type inequality and the continuity method.展开更多
This study investigated the impact of the xu-argument-based continuation on Chinese high school students’English syntactic complexity captured using verb-argument constructions(VACs)over an 8-week period.Participants...This study investigated the impact of the xu-argument-based continuation on Chinese high school students’English syntactic complexity captured using verb-argument constructions(VACs)over an 8-week period.Participants were two comparable groups of students:one group worked with English input texts(i.e.,E-E),while the other worked with Chinese input texts with the same content(i.e.,C-E).The results showed that over time,the E-E group exhibited a greater tendency to use a wider range of VACs,such as caused-motion constructions,attributives,passives,and phrasal verbs.At the same time,they reduced their use of simpler VACs like intransitive-motion and simple transitive constructions,especially when compared to the C-E group.This pattern was also evident in the topic-based writing during the posttest.These findings strongly support the effectiveness of xu-argument-based continuation tasks in promoting the development of L2 VAC knowledge.They suggest that tasks combining language input with output can significantly enhance learners’ability to use more sophisticated VACs.展开更多
In this paper,we propose a learning algorithm termed linear multistep adaptive moment(LMAdam) to enhance the adaptive moment(Adam) algorithm for machine learning.Considering Adam as a single-step discretization of its...In this paper,we propose a learning algorithm termed linear multistep adaptive moment(LMAdam) to enhance the adaptive moment(Adam) algorithm for machine learning.Considering Adam as a single-step discretization of its continuous counterpart,we develop the LMAdam algorithm based on a linear multistep discretization scheme.We design a feedforward neural network for learning the coefficients of the multistep terms with ensured consistency and select the coefficients to ensure zero stability of the multistep terms.We experimentally demonstrate the superiority of the LMAdam via extensive experimentation on benchmark datasets for training various deep neural networks in three applications.展开更多
This study compares the relative efficacy of the continuation task and the model-as-feedbackwriting (MAFW) task in EFL writing development. Ninety intermediate-level Chinese EFL learnerswere randomly assigned to a con...This study compares the relative efficacy of the continuation task and the model-as-feedbackwriting (MAFW) task in EFL writing development. Ninety intermediate-level Chinese EFL learnerswere randomly assigned to a continuation group, a MAFW group, and a control group, each with30 learners. A pretest and a posttest were used to gauge L2 writing development. Results showedthat the continuation task outperformed the MAFW task not only in enhancing the overall qualityof L2 writing, but also in promoting the quality of three components of L2 writing, namely, content,organization, and language. The finding has important implications for L2 writing teaching andlearning.展开更多
The flow behavior of molten steel in the thin slab mold under high casting speed conditions was investigated,with a focus on the multi-mode continuous casting and rolling mold.A steel-slag two-phase flow model was est...The flow behavior of molten steel in the thin slab mold under high casting speed conditions was investigated,with a focus on the multi-mode continuous casting and rolling mold.A steel-slag two-phase flow model was established using large eddy simulation,the volume of fluid,and magnetohydrodynamics methods through numerical simulation.The maximum flow velocity and wave height at the steel-slag interface within the mold are critical evaluation criteria for analyzing asymmetric flow under varying casting speeds and electromagnetic braking.The results indicate that the asymmetric flows within the mold do not occur synchronously.The severity of the asymmetric flow correlates with the velocity difference across the steel-slag interface.A greater biased flow prolongs the time required to revert to a steady state.When the magnetic field intensity is set to 0.24 T and the magnetic pole position is at 390 mm from the steel-slag interface,this configuration can reduce the velocity of the steel-slag interface,thereby mitigating the asymmetric flow.Additionally,it can diminish the velocity,impact depth,and impact intensity on the narrow face of the jet,thus improving the distribution of velocity and turbulent kinetic energy within the mold.This configuration prolongs the time required for the steel-slag interface to transition from a stable state to its maximum velocity and shortens the time for the interface to return to stability from an unstable state.Moreover,it ensures the positional stability of the steel-slag interface,confining its position within−3 mm.展开更多
文摘This study examines the reliability and validity of AI-generated scoring for continuation writing tasks.By comparing GPT-4 with eight experienced human raters across 21 student responses,it evaluates AI’s consistency,severity,and alignment with human scoring criteria.Results show that AI exhibits high self-consistency and adapts effectively to different scoring roles(e.g.,teacher vs.highstakes rater).However,AI scores were more lenient than human raters and demonstrated divergent evaluation focuses—prioritizing narrative coherence and emotional depth,while teachers emphasized linguistic accuracy and richness of detail.The findings suggest AI’s potential as a supplementary assessment tool,offering rapid,holistic feedback,but highlight the need for further calibration to align with educational standards.Implications include exploring hybrid evaluation models that leverage the strengths of both AI and human raters to achieve more equitable,efficient,and pedagogically meaningful writing assessments.
文摘This study investigated the effects of xu-argument-based continuation writing on learners’processing of source texts.Seventy-five participants were randomly assigned to three conditions:(1)continuation writing,(2)summary writing,or(3)reading comprehension.Eye-tracking data were collected during reading,measuring early(first fixation duration,first pass duration)and late(go-past time,total fixation duration)eye movements.During writing,source-text rereading was tracked via fixation counts and durations.Results showed that task type did not affect initial lexical access,as first fixation duration showed no group differences.However,both production groups exhibited significantly longer first pass durations than the reading comprehension group.Late measures revealed a gradient pattern:the continuation writing group spent significantly longer gopast time and total fixation duration than the summary writing group,which exceeded the reading comprehension group.This indicates that continuation tasks promoted deeper cognitive engagement during reading.During writing,the continuation writing group spent more time rereading the source text with higher fixation counts than the summary writing group.These findings suggest that continuation writing triggers more intensive reader-text interaction during pre-writing and enhances comprehension-production coupling through sustained attention to input during writing.This study sheds light on the cognitive mechanisms underlying the theoretical and pedagogical value of xu-argument.
基金funded by the Medical Science Research Project of the Hebei Provincial Health and Health Commission(20230550).
文摘BACKGROUND:This study is to evaluate clearance eff ects of hemoperfusion(HP),continuous renal replacement therapy(CRRT),and plasma exchange(PE)for chlorfenapyr and its metabolite tralopyril in patients with acute poisoning.METHODS:This retrospective study included 18 patients with acute oral chlorfenapyr poisoning treated at our department between January 2022 and January 2024.All patients received conventional therapies combined with blood purification,including HP,CRRT,and PE.HP was performed three sessions within the fi rst 24 h,followed by CRRT and PE.Serial blood samples were collected to measure plasma concentrations of chlorfenapyr and tralopyril using gas chromatography/liquid chromatography-mass spectrometry(GC/LC-MS).The toxin-clearance eff ects were assessed using a linear mixed-eff ects(LME)model.RESULTS:The hourly decline rate of the plasma chlorfenapyr concentration(median[IQR])was 8.83%(1.79%)for HP,4.12%(1.26%)for CRRT,and 6.85%(1.44%)for PE.LME analysis showed higher decline rate in the plasma concentration with HP(β=5.00;P<0.001)and PE(β=2.15;P=0.003)compared to CRRT.For tralopyril,the hourly decline rates were 3.04%(0.62%)for HP,1.82%(0.48%)for CRRT,and 3.01%(0.37%)for PE.LME analysis showed that the clearance effects of HP(β=0.027;P<0.001)and PE(β=0.022;P=0.001)were superior to CRRT.Pre-treatment toxin levels and the interval from hospital admission to blood purifi cation showed no signifi cant interaction with clearance outcomes.CONCLUSION:In our study,HP was associated with a higher decline rate in plasma chlorfenapyr concentration compared to CRRT and PE,supporting HP as a preferred early intervention.However,all three methods showed limited effi cacy in reducing tralopyril levels.Further research into the toxicokinetics and mechanisms of chlorfenapyr is warranted to optimize purifi cation strategies.
文摘Background:Continuous care for children with enterostomy and their families has been gaining popularity in China.Objective:To evaluate the feasibility of continuous care for children with enterostomy and their families in China.Methods:The PubMed,Web of Science,Embase,Cochrane Library,EBSCO,CNKI,CBM,VIP,and WanFang were searched for clinical trials until December 30,2025.Two reviewers independently searched articles,evaluated quality and extracted data.This review was conducted in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses(PRISMA).Results:33 studies involving 2774 participants were included.The meta-analysis showed that continuous care strategy can significantly reduce the incidence of complications in children with enterostomy(OR=0.20,95%CI=0.16-0.26,p<0.001,I2=0%),effectively improve the family caregiver ability for enterostomy(MD=-10.34,95%CI=-13.82 to-6.85,p<0.001,I2=99%),shorten the time for family members to replace stoma bags(MD=-13.57,95%CI=-19.66 to-7.49,p<0.001,I2=100%),and alleviate negative emotions such as anxiety(SMD=-1.80,95%CI=-2.36 to-1.23,p<0.001,I2=92%)and depression(SMD=-1.54,95%CI=-2.04 to-1.04,p<0.001,I2=89%)in the families of the affected children.Conclusions:Continuous care can reduce complications of enterostomy in children,improve the family caregiver ability for enterostomy and alleviate negative emotions of family members such as anxiety and depression.
基金Project supported by the National Natural Science Foundation of China(No.12572020)the Key Project of Natural Science Foundation of Hebei Province of China(No.A2023210064)。
文摘Local resonant acoustic metamaterials have broad applications in sound insulation,yet their single-configuration designs often exhibit limited and discontinuous bandgap widths,hindering full-frequency noise attenuation across the human auditory range.This study presents a double-phase fidget-spinner-shaped acoustic metamaterial(DFAM),specifically designed to achieve an ultra-broad,low-frequency continuous bandgap by means of synergistic structural optimization,enabling effective and robust control of audible noise.Based on Bloch's theorem and the finite element method,the dispersion relation of the DFAM structure is calculated and verified by the transmission loss curves.The propagation characteristics of sound waves within the structure are further analyzed for noise frequencies that fall within the passband.The influence of the geometric and physical parameters on the bandgap is investigated,and the corresponding transmission loss in the propagation direction is further calculated.A hybrid collaborative design strategy,leveraging multi-parameter optimization and bandgap complementarity,is developed to construct a metastructure with continuous bandgap coverage from 20 Hz to 1000 Hz.The resulting metastructure demonstrates exceptional broadband noise attenuation,achieving a total bandgap width of 876.3 Hz(87.63% of the target range)with the transmission loss up to-762.78 d B in a three-periodic arrangement.The simulation and experimental results for the transmission loss of the DFAM metastructure show strong agreement in the low-frequency range.This work provides a novel framework for designing ultra-wide low-frequency continuous bandgap metastructures,offering significant potential for noise mitigation in complex environments.
基金supported by the National Key Research and Development Program of China(Grant No.2021YFB3700602)the Jiaxing Key Research and Development Program(Grant No.2022BZ10010).
文摘The refinement of the as-cast grain structure in austenitic heat-resistant stainless steel depends on the formation of active solid nuclei during solidification.Titanium(Ti)additions successfully induced the formation of Ti-containing inclusions,enhancing heterogeneous nucleation and promoting equiaxed dendritic growth in 347H stainless steel.Thermal simulation experiments indicated that the equiaxed crystal ratios increased notably with Ti content;samples with 0.06,0.12,and 0.36 wt.%Ti exhibited equiaxed ratios of 18%,24%,and 41%,respectively.Three primary inclusion types—TiN,Al_(2)O_(3)-TiN,and TiO_(x)-TiN—were identified at the cores of equiaxed dendrites,with nucleation core sizes predominantly ranging from 0.5 to 8μm.Among the tested samples,the 0.36 wt.%Ti addition produced the highest nucleation core density.Increasing Ti content significantly elevated dendrite tip undercooling from 2.6 K(0.06 wt.%Ti)to 10.8 K(0.36 wt.%Ti),accelerating solidification front instability and thus enhancing heterogeneous nucleation.Additionally,higher Ti content increased the divergence angle between adjacent columnar dendrites,further promoting the columnar-to-equiaxed transition(CET).
文摘Introduction:The choice is no longer whether—but when and how!Today,organizations can no longer choose whether or not to engage with stakeholders;the only real decision is when and how to do it successfully.Engaging stakeholders is essential for any type of organization,public,private,or civil society.In fact,it is a fundamental part of effective governance,continuous improvement,and social responsibility.The principle behind stakeholder engagement is simple yet powerful:those who can influence or be affected by an organization's mission must be given the opportunity to express their views and contribute to shaping the decisions that impact them.
基金Supported by National Natural Science Foundation of China(Grant No.52272440)Suzhou Science Foundation(Grant Nos.SYG202323,ZXL2022027).
文摘Continual learning fault diagnosis(CLFD)has gained growing interest in mechanical systems for its ability to accumulate and transfer knowledge in dynamic fault diagnosis scenarios.However,existing CLFD methods typically assume balanced task distributions,neglecting the long-tailed nature of real-world fault occurrences,where certain faults dominate while others are rare.Due to the long-tailed distribution among different me-chanical conditions,excessive attention has been focused on the dominant type,leading to performance de-gradation in rarer types.In this paper,decoupling incremental classifier and representation learning(DICRL)is proposed to address the dual challenges of catastrophic forgetting introduced by incremental tasks and the bias in long-tailed CLFD(LT-CLFD).The core innovation lies in the structural decoupling of incremental classifier learning and representation learning.An instance-balanced sampling strategy is employed to learn more dis-criminative deep representations from the exemplars selected by the herding algorithm and new data.Then,the previous classifiers are frozen to prevent damage to representation learning during backward propagation.Cosine normalization classifier with learnable weight scaling is trained using a class-balanced sampling strategy to enhance classification accuracy.Experimental results demonstrate that DICRL outperforms existing continual learning methods across multiple benchmarks,demonstrating superior performance and robustness in both LT-CLFD and conventional CLFD.DICRL effectively tackles both catastrophic forgetting and long-tailed distribution in CLFD,enabling more reliable fault diagnosis in industrial applications.
基金the Science and Technology R&D Major Project of Jiangxi Province(No.20244AFI92001)the National Natural Science Foundation of China(Nos.22071033 and 21801047)for the financial supports.
文摘We report an immobilized enzyme-catalyzed batch and continuous-flow synthesis of optically pure ethyl(R)-pantothenate((R)-PaOEt),the direct precursor of d-pantothenic acid.Firstly,a ketoreductase mutant designated as M2,carrying two-point mutations of F97L and M242F relative to the wild-type SSCR,was constructed by site-directed mutagenesis,exhibited simultaneously improved activity toward ethyl 2′-ketopantothenate(K-PaOEt)and isopropanol,and could effectively catalyze the stereoselective reduction of K-PaOEt to(R)-PaOEt by using isopropanol as the sacrificial co-substrate to regenerate NADPH.After screening six commercially available carriers,an amino resin LXTE-700 was identified as the best solid support for the immobilization of M2 via the glutaraldehyde activation method.Upon optimization of the immobilization process and reaction conditions,the fabricated immobilized enzyme M2@amino resin demonstrated excellent recyclability and reusability,with the complete conversion of K-PaOEt to(R)-PaOEt being still realized after 12 cycles of reuse.Finally,M2@amino resin-catalyzed synthesis of(R)-PaOEt was successfully implemented in continuous-flow,accomplishing a 6.3 times higher space-time yield than that with the batch synthesis(529.2 versus 84 g L^(-1) d^(-1)).Our developed flow biocatalysis system also features an outstanding operational stability,as evidenced by the 100%conversion rate achieved after 15 consecutive days of operation.
文摘1.Introduction.Pancreaticoduodenectomy is an essential surgical procedure for the treatment of malignant tumors in the pancreatic head,distal common bile duct,and duodenal papilla,and is widely used in clinical practice.One of the primary determinants of surgical success is the durability of the pancreaticojejunal anastomosis.
文摘The iterative continuation task(ICT)requires English as a foreign language(EFL)learners to read a segment and write a continuation that aligns with the preceding segment of an English novel with successive turns,offering exposure to diverse grammatical structures and opportunities for contextualized usage.Given the importance of integrating technology into second language(L2)writing and the critical role that grammar plays in L2 writing development,automated written corrective feedback provided by Grammarly has gained significant attention.This study investigates the impact of Grammarly on grammar learning strategies,grammar grit,and grammar competence among EFL college students engaged in ICT.This study employed a mixed-methods sequential exploratory design;56 participants were divided into an experimental group(n=28),receiving Grammarly feedback for ICT,and a control group(n=28),completing ICT without Grammarly feedback.Quantitative results revealed that both groups showed improvements in L2 grammar learning strategies,grit and competence.For the experimental group,significant differences were observed across all variables of L2 grammar learning strategies,grit,and competence between pre-and post-tests.For the control group,significant differences were only observed in the affective dimension of grammar learning strategies,Consistency of Interest(COI)of grammar grit,and grammar competence.However,the control group presented a significantly higher improvement in grammar competence.Qualitative analysis showed both positive and negative perceptions of Grammarly.The pedagogical implications of integrating Grammarly and ICT for L2 grammar development are discussed.
基金supported by the National Natural Science Foundation of China(Nos.52422408 and 52171031)the Liaoning Xingliao Talents-Top-notch Young Talents Project(No.XLYC2203064)+1 种基金the Excellent Youth Fund of Liaoning Natural Science Foundation(No.2023JH3/10200001)the Fundamental Research Funds for the Central Universities(No.N2425004).
文摘The transient phenomena of re-oxidation and slag entrapment occurring in the tundish during the ladle change-over process have been proven detrimental to clean steel production.Therefore,an unsteady three-phase turbulence model,coupling velocity,temperature,and phase field was established to study the effect of the ladle shroud immersion depth on the slag eye formation,slag entrainment,slag dragging,air dragging,and flow characteristics during the ladle change-over process of a two-strand tundish.The results showed that reducing the immersion depth decreases the high-velocity region area under the slag layer in the quasi-steady process.During the emptying stage,as the molten bath level gradually decreases,the outlet temperature exhibits a trend of initially decreasing and subsequently increasing across all three shroud immersion depths.However,under a 210 mm shroud immersion depth,molten slag and air are dragged into the shroud,forming slag droplets and causing significant fluctuations,with a maximum scalar velocity of 0.0764 m/s at the monitoring point.In the filling stage,air and molten slag are dragged into the molten bath,forming bubbles and slag droplets at an immersion depth of 210 mm.Bubbles are observed within the molten slag layer,which can readily cause an emulsification phenomenon,making it easier to be dragged as slag droplets.Additionally,the slag eye area measured under 210 mm immersion depth at 45 s is 0.303 m^(2),while the maximum scalar velocity of 2.4259 m/s is detected at 12 s.At an immersion depth of 360 mm,the average area of the slag eye is minimized to 0.06268 m2,with corresponding variances of 0.006753,representing the optimal immersion depth.
基金supported by the National Natural Science Foundation of China(No.52274318).
文摘A full-sectional microstructure characterization method was developed to investigate the formation of coarse slag rims during the continuous casting of hypo-peritectic steel.The cross-sectional microstructural analysis of typical slag rims for two highly crystalline powders revealed that their formation was primarily driven by the solidification of the liquid slag.Distinct differences were observed in the microstructures of slag rims from the two powders.Powder A(characterized by a higher breaking temperature and viscosity)displayed alternating lamellar microstructures of coarse and fine phases,with the coarse phases composed of akermanite-gehlenite transition phases.In contrast,powder B(with a lower breaking temperature and viscosity)predominantly comprised regular akermanite-gehlenite crystals interspersed with a certain amount of glassy phases.Numerical simulations of a three-phase fluid flow coupled with heat transfer indicate that slag rim formation correlates with mold oscillation.Solidification of the liquid slag at the slag rim front predominantly occurs during the negative stroke of the mold oscillation.The average heating rate during the ascending stage of the mold reaches approximately 100 K·s^(−1),whereas the average cooling rate during the descending stage attains 400 K·s^(−1).This temperature variation leads to the formation of lamellar microstructures,whereas the ascending stage promotes the formation of coarse structures and thicker slag rims.Based on the powder properties,two distinct formation pathways exist for highly crystalline mold powders.For the powders with a higher breaking temperature,higher viscosity,and narrower solidification range(powder A),coarse microstructures and thicker slag rims were preferentially formed.For powders with lower breaking temperature and viscosity and wider solidification ranges(powder B),the liquid slag resisted rapid solidification,and the extended mushy zone allowed the partial liquid slag to persist at the slag rim front,promoting the formation of a thin slag rim.This study enhances the understanding of slag rim formation in highly crystalline mold powders and provides critical insights into the control of longitudinal surface cracks in hypo-peritectic steel.
文摘This study integrates explicit input enhancement into comparative continuation writing,defined as a task in which learners produce a continuation by comparing their own expression with an input text,aligning with its discourse structure and linguistic features,while developing their own ideas.It aims to examine whether English as a Foreign Language(EFL)learners in China exhibit differences in discourse competence and writing performance when completing comparative continuation writing combined with different input enhancement techniques,and whether the alignment effect occurs at the discourse level.Sixty first-year Chinese senior middle school students were divided into four groups:three groups engaged in comparative continuation writing with varying input enhancement,achieved by combining different techniques,while a control group performed a designated-topic writing task.The results revealed that three comparative continuation writing groups outperformed the designated-topic writing group in discourse competence,particularly in the use of temporal connectives.However,differences and some inconsistencies were observed among the comparative continuation writing groups across individual indices.The study highlights effective ways to incorporate comparative continuation writing into English instruction and demonstrates how explicit input enhancement can complement the task,simultaneously activating the alignment effect proposed by the xu-argument and enhancing discourse competence in writing.
基金supported by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(RS-2025-00518960)in part by the National Research Foundation of Korea(NRF)grant funded by the Korea government(MSIT)(RS-2025-00563192).
文摘Mental-health risk detection seeks early signs of distress from social media posts and clinical transcripts to enable timely intervention before crises.When such risks go undetected,consequences can escalate to self-harm,long-term disability,reduced productivity,and significant societal and economic burden.Despite recent advances,detecting risk from online text remains challenging due to heterogeneous language,evolving semantics,and the sequential emergence of new datasets.Effective solutions must encode clinically meaningful cues,reason about causal relations,and adapt to new domains without forgetting prior knowledge.To address these challenges,this paper presents a Continual Neuro-Symbolic Graph Learning(CNSGL)framework that unifies symbolic reasoning,causal inference,and continual learning within a single architecture.Each post is represented as a symbolic graph linking clinically relevant tags to textual content,enriched with causal edges derived from directional Point-wise Mutual Information(PMI).A two-layer Graph Convolutional Network(GCN)encodes these graphs,and a Transformer-based attention pooler aggregates node embeddings while providing interpretable tag-level importances.Continual adaptation across datasets is achieved through the Multi-Head Freeze(MH-Freeze)strategy,which freezes a shared encoder and incrementally trains lightweight task-specific heads(small classifiers attached to the shared embedding).Experimental evaluations across six diverse mental-health datasets ranging from Reddit discourse to clinical interviews,demonstrate that MH-Freeze consistently outperforms existing continual-learning baselines in both discriminative accuracy and calibration reliability.Across six datasets,MH-Freeze achieves up to 0.925 accuracy and 0.923 F1-Score,with AUPRC≥0.934 and AUROC≥0.942,consistently surpassing all continual-learning baselines.The results confirm the framework’s ability to preserve prior knowledge,adapt to domain shifts,and maintain causal interpretability,establishing CNSGL as a promising step toward robust,explainable,and lifelong mental-health risk assessment.
基金Supported by National Natural Science Foundation of China(Grant No.11601122).
文摘The paper considers the initial value problem of inhomogeneous fourth-order Schr¨odinger equation with potential in energy space H^(2)(R^(d)).The global well-posedness is obtained in dimensions d≥5 resorting to contractive mapping principle,Strichartz estimates,Caffarelli-Kohn-Nirenberg-type inequality and the continuity method.
文摘This study investigated the impact of the xu-argument-based continuation on Chinese high school students’English syntactic complexity captured using verb-argument constructions(VACs)over an 8-week period.Participants were two comparable groups of students:one group worked with English input texts(i.e.,E-E),while the other worked with Chinese input texts with the same content(i.e.,C-E).The results showed that over time,the E-E group exhibited a greater tendency to use a wider range of VACs,such as caused-motion constructions,attributives,passives,and phrasal verbs.At the same time,they reduced their use of simpler VACs like intransitive-motion and simple transitive constructions,especially when compared to the C-E group.This pattern was also evident in the topic-based writing during the posttest.These findings strongly support the effectiveness of xu-argument-based continuation tasks in promoting the development of L2 VAC knowledge.They suggest that tasks combining language input with output can significantly enhance learners’ability to use more sophisticated VACs.
基金supported in part by the National Natural Science Foundation of China(62506148 and 62476115)the Fundamental Research Funds for the Central Universities(lzujbky-2025-pd05 and lzujbky-2025-ytB01)+2 种基金the Research Grants Council of the Hong Kong Special Administrative Region of China(AoE/E-407/24-N and C1013-24G)the Postdoctoral Fellowship Program(Grade C) of China Postdoctoral Science Foundation(GZC20251039)the Supercomputing Center of Lanzhou University。
文摘In this paper,we propose a learning algorithm termed linear multistep adaptive moment(LMAdam) to enhance the adaptive moment(Adam) algorithm for machine learning.Considering Adam as a single-step discretization of its continuous counterpart,we develop the LMAdam algorithm based on a linear multistep discretization scheme.We design a feedforward neural network for learning the coefficients of the multistep terms with ensured consistency and select the coefficients to ensure zero stability of the multistep terms.We experimentally demonstrate the superiority of the LMAdam via extensive experimentation on benchmark datasets for training various deep neural networks in three applications.
文摘This study compares the relative efficacy of the continuation task and the model-as-feedbackwriting (MAFW) task in EFL writing development. Ninety intermediate-level Chinese EFL learnerswere randomly assigned to a continuation group, a MAFW group, and a control group, each with30 learners. A pretest and a posttest were used to gauge L2 writing development. Results showedthat the continuation task outperformed the MAFW task not only in enhancing the overall qualityof L2 writing, but also in promoting the quality of three components of L2 writing, namely, content,organization, and language. The finding has important implications for L2 writing teaching andlearning.
基金support from the National Natural Science Foundation of China(Grant Nos.52174313 and 52304350)thank all members of the Hebei High Quality Steel Continuous Casting Engineering Technology Research Center at North China University of Science and Technology,Tangshan,China.
文摘The flow behavior of molten steel in the thin slab mold under high casting speed conditions was investigated,with a focus on the multi-mode continuous casting and rolling mold.A steel-slag two-phase flow model was established using large eddy simulation,the volume of fluid,and magnetohydrodynamics methods through numerical simulation.The maximum flow velocity and wave height at the steel-slag interface within the mold are critical evaluation criteria for analyzing asymmetric flow under varying casting speeds and electromagnetic braking.The results indicate that the asymmetric flows within the mold do not occur synchronously.The severity of the asymmetric flow correlates with the velocity difference across the steel-slag interface.A greater biased flow prolongs the time required to revert to a steady state.When the magnetic field intensity is set to 0.24 T and the magnetic pole position is at 390 mm from the steel-slag interface,this configuration can reduce the velocity of the steel-slag interface,thereby mitigating the asymmetric flow.Additionally,it can diminish the velocity,impact depth,and impact intensity on the narrow face of the jet,thus improving the distribution of velocity and turbulent kinetic energy within the mold.This configuration prolongs the time required for the steel-slag interface to transition from a stable state to its maximum velocity and shortens the time for the interface to return to stability from an unstable state.Moreover,it ensures the positional stability of the steel-slag interface,confining its position within−3 mm.
