Theintegration of human factors into artificial intelligence(AI)systems has emerged as a critical research frontier,particularly in reinforcement learning(RL),where human-AI interaction(HAII)presents both opportunitie...Theintegration of human factors into artificial intelligence(AI)systems has emerged as a critical research frontier,particularly in reinforcement learning(RL),where human-AI interaction(HAII)presents both opportunities and challenges.As RL continues to demonstrate remarkable success in model-free and partially observable environments,its real-world deployment increasingly requires effective collaboration with human operators and stakeholders.This article systematically examines HAII techniques in RL through both theoretical analysis and practical case studies.We establish a conceptual framework built upon three fundamental pillars of effective human-AI collaboration:computational trust modeling,system usability,and decision understandability.Our comprehensive review organizes HAII methods into five key categories:(1)learning from human feedback,including various shaping approaches;(2)learning from human demonstration through inverse RL and imitation learning;(3)shared autonomy architectures for dynamic control allocation;(4)human-in-the-loop querying strategies for active learning;and(5)explainable RL techniques for interpretable policy generation.Recent state-of-the-art works are critically reviewed,with particular emphasis on advances incorporating large language models in human-AI interaction research.To illustrate some concepts,we present three detailed case studies:an empirical trust model for farmers adopting AI-driven agricultural management systems,the implementation of ethical constraints in roboticmotion planning through human-guided RL,and an experimental investigation of human trust dynamics using a multi-armed bandit paradigm.These applications demonstrate how HAII principles can enhance RL systems’practical utility while bridging the gap between theoretical RL and real-world human-centered applications,ultimately contributing to more deployable and socially beneficial intelligent systems.展开更多
数智时代,用户与人工智能系统的交互模式正在发生怎样的转变?《Human-AI Interaction and Collaboration》一书通过理论框架构建、应用场景描述、潜在风险辨识,为读者全方位地解析了人智交互与协作的内在逻辑,提出“以人为本”这一人工...数智时代,用户与人工智能系统的交互模式正在发生怎样的转变?《Human-AI Interaction and Collaboration》一书通过理论框架构建、应用场景描述、潜在风险辨识,为读者全方位地解析了人智交互与协作的内在逻辑,提出“以人为本”这一人工智能系统设计的首要原则,详细分析了用户感知、系统设计、人智关系等一系列因素如何影响人智协作的效率与成果,在针对潜在风险制定应对方案的同时,为读者展示了人智协作在医疗、科研、金融等多学科场景下的广阔前景。该书为未来人智交互与协作相关研究提供了坚实的理论框架,也为人工智能系统设计者提供了伦理指南。展开更多
Interactional naturalness of human-AI dialogue is a key factor in fostering meaningful communicative engagement.This study addresses this gap by adopting Conversation Analysis(CA)as a methodological framework to exami...Interactional naturalness of human-AI dialogue is a key factor in fostering meaningful communicative engagement.This study addresses this gap by adopting Conversation Analysis(CA)as a methodological framework to examine recorded speaking practice sessions between English as a Second or Foreign Language(ESL/EFL)learners and an AI tutoring system.Drawing on the principles of emergence,indexicality,and recipient design,the study investigates why and how interactional naturalness is violated in AI-mediated conversation.These findings reveal that while the AI tutor produces grammatically well-formed and topically relevant utterances,its conduct frequently diverges from the moment-by-moment,context-sensitive organization that characterizes natural human interaction.Particularly,the AI system fails to consistently align with emergent turn sequences,indexical references,and recipient-designed responses.The results indicate that current AI tutoring systems,despite advanced language models,have not yet achieved interactional naturalness as defined from a conversation-analytic perspective.Therefore,a shift in evaluation criteria from technical accuracy to interactional quality in the design and assessment of AI-based language tutors is called-for.展开更多
As a multidisciplinary phenomenon,panel aeroelasticity in shock-dominated flow is featured by two primary interactions:Fluid-Structure Interactions(FSIs)and Shock-Boundary Layer Interactions(SBLIs).The former raises s...As a multidisciplinary phenomenon,panel aeroelasticity in shock-dominated flow is featured by two primary interactions:Fluid-Structure Interactions(FSIs)and Shock-Boundary Layer Interactions(SBLIs).The former raises structural concerns,and the latter is of aerodynamic interest.Thus,panel aeroelasticity in shock-dominated flow represents a vital topic for the development and optimization of supersonic vehicles and propulsion systems.This review systematically summarizes recent advances in the methodologies applied to capture structural and fluid dynamics,including theoretical models,numerical simulations,and wind tunnel experiments.The application of data-driven modal decomposition,an advanced technique to extract physically crucial features,on the topic is introduced.From the perspective of FSIs,the distinctive aeroelastic behaviors in shock-dominated flow,including hysteresis phenomena and nonlinear responses,are highlighted.From the perspective of SBLIs,the modifications in their spatial and temporal characteristics imposed by the aeroelastic responses are emphasized.Motivated by the interaction between the shock waves and structural response,different strategies have been proposed to implement aeroelastic suppression and shock control,which have the potential to enhance structural safety and aerodynamic performance in the next generation of high-speed flight vehicles.展开更多
The characterization of drug-target interactions is a key component of drug discovery,testing,and development.Affinity chromatography is one approach that can be used for this type of analysis.For instance,this may be...The characterization of drug-target interactions is a key component of drug discovery,testing,and development.Affinity chromatography is one approach that can be used for this type of analysis.For instance,this may be done by using an immobilized target as a stationary phase and a drug as the applied solute.This review will discuss the various ways in which affinity chromatographic methods have been used to examine drug-target interactions,with an emphasis on high-performance methods.The general principles of this approach and factors to consider in its use for drug-target interaction analysis will first be examined.Methods based on zonal elution or frontal analysis for binding and competition studies will then be discussed.Various techniques for kinetic studies will next be considered,along with approaches that employ secondary binding agents and hybrid techniques.In each case,the general principles and theory of an approach will be given along with examples of its use in drug-target interaction studies.Advantages or limitations of each approach will be provided as well.This information should make it possible in the future to extend these techniques to other drug-target systems of interest in biomedical research and drug testing or development.展开更多
Affinity selection mass spectrometry(AS-MS)has emerged as a powerful label-free technique for identifying and characterizing ligand-target interactions.This review explores the diverse applications of AS-MS in drug di...Affinity selection mass spectrometry(AS-MS)has emerged as a powerful label-free technique for identifying and characterizing ligand-target interactions.This review explores the diverse applications of AS-MS in drug discovery,including its role in selective screening,binding site characterization,and quantitative affinity determination.We discuss the use of AS-MS for determining equilibrium dissociation constants(KD)and competitive binding parameters(affinity competition experiment 50%(ACE50)),highlighting its ability to rank ligand affinities efficiently.The review also examines AS-MS applications in fragment-based drug discovery(FBDD),screening for molecular glues,and investigating interactions with membrane proteins.Moreover,we address key technical challenges,including competitive binding effects,protein stability,and ligand dissociation kinetics,along with recent advancements in automation and artificial intelligence(AI)integration.Rather than providing a comprehensive literature review,this work aims to broaden the applicability of AS-MS assays and encourage researchers to explore its use in underutilized contexts.By providing rapid and high-sensitivity affinity measurements,AS-MS continues to expand its role in drug discovery and structural biology,complementing conventional biophysical techniques.展开更多
Structure-ice interaction problems have attracted increasing attention,yet accurately predicting the loads exerted by sea ice on ship hulls remains a significant challenge.Over the past few decades,various numerical m...Structure-ice interaction problems have attracted increasing attention,yet accurately predicting the loads exerted by sea ice on ship hulls remains a significant challenge.Over the past few decades,various numerical methods have been employed to simulate ice resistance on ships and evaluate their manoeuvrability during ice-structure interaction.Among these approaches,the circumferential crack method has demonstrated both high efficiency and accuracy.This paper provides a detailed introduction to the fundamental theory of this method,including the numerical modeling of different failure modes and the dynamic ice motion responses.Furthermore,it reviews existing studies on predicting ice resistance and assessing the manoeuvrability of icebreakers navigating through ice-covered regions using the circumferential crack method.Several recommendations for future research in this field are also presented.展开更多
Research in human‑robot Interaction(HRI)has increasingly demonstrated how Augmented Reality(AR)enables better interactions between humans and robots.However,the design of HRI remains less understood.Through a systemat...Research in human‑robot Interaction(HRI)has increasingly demonstrated how Augmented Reality(AR)enables better interactions between humans and robots.However,the design of HRI remains less understood.Through a systematic literature review of 53 related papers,this research provides an overview of the emerging applications and trends for AR and identifies three types of AR interfaces as follows:1)remote modular interface,2)proximal modular interface,and 3)proximal integral interface.The review indicates potential future directions of construction‑oriented and human‑centric interaction design studies,leading to four pairs of subsystems,which are frequently modularised or integrated,and three conceptual frameworks for HRI interfaces are proposed.Moreover,this research contributes to the theoretical exploration of interaction design.Future applications can adapt to various tasks by using the proposed three conceptual frameworks for interfaces,as well as combining the four proposed subsystem pairs to suit specific task requirements in the construction sector.展开更多
As a common electronic adhesive,ultraviolet(UV)curing polyurethane acrylate adhesive has both flexibility and wear resistance of polyurethane,excellent weather resistance and optical properties of acrylate.Despite the...As a common electronic adhesive,ultraviolet(UV)curing polyurethane acrylate adhesive has both flexibility and wear resistance of polyurethane,excellent weather resistance and optical properties of acrylate.Despite the extensive applications,it is still difficult to solve the problems caused by the shrinkage of adhesive.Here,a new type of photosensitive adhesive for bonding electronic components based on supramolecular interaction was designed and synthesized.The supramolecular interaction of cyclodextrin and adamantane moieties introduced into the adhesive polymer entitles the viscosity of the adhesive to rise rapidly during use,thereby preventing adhesive loss and dislocation of electronic components.UV light could further cure the adhesive and position the electronic components.The adhesive shrunk<2%when cured by UV light,so it can be used for electronic packaging and high-resolution,defect-free lithography.展开更多
This study aimed to gain insight into the interaction and metabolic mechanism between Aspergillus flavus U-16 and Saccharomyces cerevisiae HJ during co-fermentation.Despite the suppressed transcriptional activity of S...This study aimed to gain insight into the interaction and metabolic mechanism between Aspergillus flavus U-16 and Saccharomyces cerevisiae HJ during co-fermentation.Despite the suppressed transcriptional activity of SU-16,its metabolites exerted a pivotal regulatory influence.The metabolites produced by SU-16 significantly increased the ethanol content to 17.0%.They also significantly increased the contents of crucial flavor substances,including amino acids(at least 5 times)and esters(up to 1100 mg/L).SU-16 produced 130 metabolites that may affect flavor compounds,while HJ mainly controlled the synthesis of 161 compounds.Metabolites of SU-16 with different molecular weights were observed to exert differential regulatory effects on the fermentation process and the metabolic pathway of HJ.The small molecule metabolites of SU-16 promoted the growth of HJ significantly.In contrast,its macromolecular metabolites were found to mainly affect the expression of genes related to carbohydrate and amino acid metabolism in the HJ genome.展开更多
Schizophrenia(SCZ)is a severe mental illness influenced by gene-environment interactions(GEI).However,little is known about how GEI mediates SCZ.The present study aimed to examine the underlying mechanism of SCZ media...Schizophrenia(SCZ)is a severe mental illness influenced by gene-environment interactions(GEI).However,little is known about how GEI mediates SCZ.The present study aimed to examine the underlying mechanism of SCZ mediated by GEI.We found that a single environmental factor(two-week adolescent social isolation)or genetic factor(the heterozygous Rims1 knockout mice)did not induce SCZ-like behaviors.Interestingly,two-week adolescent social isolation successfully caused SCZ-like behaviors in heterozygous Rims1 knockout mice,which can be rescued by anti-SCZ drugs.RNA-seq analysis further revealed that synaptic vesicle-related biological processes and target genes were enriched in the hippocampus of GEI animal model mice,which was accompanied by disturbed excitatory synaptic neurotransmission.Finally,the Nrg1 gene was decreased in our RNA-seq analysis,and supplementation of Nrg1 ameliorated SCZ-like behaviors in heterozygous Rims1 socially isolated mice.Our findings establish a novel GEI animal model and offer a potential therapeutic target in the treatment of SCZ.展开更多
Objectives:Pancreatic ductal adenocarcinoma(PDAC)is a leading cause of cancer-related mortality and mainly affects older adults,who frequently experience polypharmacy.While systemic therapy may improve outcomes in sel...Objectives:Pancreatic ductal adenocarcinoma(PDAC)is a leading cause of cancer-related mortality and mainly affects older adults,who frequently experience polypharmacy.While systemic therapy may improve outcomes in selected older patients,the use of multiple drugs increases the risk of potential drug–drug interactions(pDDIs).This study aimed to evaluate the prevalence and characteristics of pDDIs in older patients with PDAC receiving first-line systemic therapy and their potential impact on clinical outcomes.Methods:We conducted a retrospective single-center study including patients aged≥75 years with PDAC who initiated first-line systemic therapy between December 2011 and January 2023.Prescription data were retrieved from institutional and outpatient pharmacy databases,and all concomitant medications were recorded.pDDIs were assessed using Intercheck Web and classified as A(minor),B(moderate),C(major but manageable),or D(contraindicated).Patients(n=140)were stratified according to the number of active pharmaceutical ingredients(APIs)at treatment initiation:20(n=44).Results:A higher number of APIs was significantly associated with an increased burden of pDDIs,including A,B,C,and D interactions(all p20 APIs showing the highest prevalence of clinically relevant interactions.Overall,70%of patients had at least one pDDI,but no significant differences were observed in 1-year overall survival,access to second-line therapy,or chemotherapy type across API groups.Conclusion:Polypharmacy significantly increases the risk of pDDIs in older patients with PDAC,but this did not translate into differences in 1-year survival or access to subsequent therapy.展开更多
Biomolecular self-assembly systems form the cornerstone of biological structures,in which the interactions between carbohydrates and proteins play a crucial role in many life processes.This has driven the development ...Biomolecular self-assembly systems form the cornerstone of biological structures,in which the interactions between carbohydrates and proteins play a crucial role in many life processes.This has driven the development of biomimetic carbohydrate-protein supramolecular assemblies(BCPSAs),which show great potential in biomedical research.We successfully constructed a novel supramolecular hybrid nanocarrier with pH responsiveness and targeting capabilities based on specific interactions between carbohydrates and proteins.This system self-assembles into nanoparticles CA@MP5 in water through the specific binding of mannose clusters on glycosylated pillar[5]arene MP5 with lectin Con A.Additionally,through a host-guest assembly strategy with the synthesized β-D-galactopyranosyl pyridine derivative G,it complexes within the MP5 cavity to form a"double-sweet"supramolecular glyco-nanoprotein(CA@MP5⊃G).This system shows high affinity for the asialoglycoprotein receptor(ASGPR)on HepG2 liver cancer cells.In vitro,the doxorubicin(DOX)-loaded micelles(DCA@MP5⊃G)release drug effectively in acidic environments and significantly inhibit HepG2 cell growth,with reduced toxicity toward normal cells.In vivo,the nanocarrier targets tumors effectively,reducing systemic toxicity and inhibiting tumor growth.This study offers a novel construction strategy for drug delivery systems based on carbohydrate-protein interactions in anti-tumor applications,which is significant for enhancing therapeutic efficacy.展开更多
Airborne microplastics(MPs)are prevalent indoors,and due to their low aerodynamic diameter,they can be inhaled,posing potential health risks.Although the toxic effects of airborne MPs have been explored using in vivo ...Airborne microplastics(MPs)are prevalent indoors,and due to their low aerodynamic diameter,they can be inhaled,posing potential health risks.Although the toxic effects of airborne MPs have been explored using in vivo and in vitro models,the interactions between MPs and cellular receptors remain understudied.In this research,dust samples from a confined space within an academic setting were investigated for microplastics(MPs)and their abundance.Further,the most prevalent microplastics were used to study receptor binding and competition interaction studies using the In-Silico method against natural agonists and antagonists of major inflammatory receptors,including the human platelet-activating factor receptor(PAFR),C-X-C motif chemokine receptor 1(CXCR1),β2-adrenergic receptor(β2-AR),and toll-like receptor 2(TLR-2).Results revealed that polyester(PE)was the predominant polymer,accounting for 23.96%of the samples.Analysis indicated that a monomer of PET,Ethylene Terephthalate(ET),exhibited a high binding affinity of-6.5 kcal/mol with the PAFR receptor.Additionally,the molecular dynamics and protein-ligand interaction study,which involves hydrogen bonding,explains the differential binding effect of ET and the control compound with the targeted receptor.The complex was formed between the ET and the receptors ofβ2-AR,CXCR1,and TLR-2,with a maximum of seven and a minimum of one hydrogen bond throughout the simulation.This research lays a foundation for understanding the potential health implications of MPs in confined office spaces,underscoring the need for further in vivo and in vitro examinations.展开更多
Amid accelerating climate change and intensifying human activities,flood risks have been accumulating,rein forcing the bidirectional interplay between humans and floods.Within this context,human-flood interaction has ...Amid accelerating climate change and intensifying human activities,flood risks have been accumulating,rein forcing the bidirectional interplay between humans and floods.Within this context,human-flood interaction has emerged as a rapidly expanding research frontier that is attracting growing attention.Here,we review 78 rep resentative studies to examine publication trends,research hotspot shifts,and persistent challenges in this field,combining bibliometric analyses with flood disaster records.We found that,over the past three decades,publica tions on human-flood interaction have increased exponentially,with their focus shifting from one-way impacts towards two-way human-flood coupling.The global collaboration network displays a“dual-core”structure,with transcontinental collaboration between China and the United States and regional collaborations within Europe,whereas participation from Global South countries remains marginal.In addition,a spatial disconnect persists between the geographic distribution of research hotspot regions and that of flood records.Based on bibliometric mapping,we further highlight three enduring challenges:(i)scale mismatches,(ii)spatiotemporal coupling gaps,and(iii)deficits in behavioral modeling.The insights gained from this study will help guide future research and investment in human-flood interaction,with practical implications for enhancing resilience to flood disasters and promoting sustainable human-water coexistence.展开更多
Hydrogels,with their hydrophilicity,flexibility,and environmental friendliness,are highly desirable for moisture-electric generators(MEGs)that harness ubiquitous moisture to generate electrical energy.As the active ma...Hydrogels,with their hydrophilicity,flexibility,and environmental friendliness,are highly desirable for moisture-electric generators(MEGs)that harness ubiquitous moisture to generate electrical energy.As the active material layer in MEGs,hydrogels play a crucial role in absorbing atmospheric moisture and converting chemical potential energy into electricity.However,the relatively low output current of the device and the instability of hydrogels pose challenges to the development of high-performance hydrogel-based MEGs.Herein,we introduce a straightforward,feasible,cost-effective,and versatile two-step solvent displacement strategy to overcome the barrier associated with the development of MEGs.Through tunable solvent interactions of glycerol and water,the moisture absorption capability and stability of the hydrogel can be improved,while promoting favorable ion migration.Such an effective processing route not only significantly boosts the output performances but also greatly improves the long-term durability of hydrogel-based MEGs.Notably,the current output and power density of the treated MEGs can increase by up to two orders of magnitude.The mechanisms behind the intriguing observation are investigated by various characterizations and theoretical calculations.This universal strategy holds promise to be extended to various hydrogel-based MEGs.Moreover,the MEGs can be used for energy harvesting,self-powered respiratory monitoring,and non-contact humidity detection.This work offers new opportunities for advancing green energy and self-powered technologies.展开更多
The cross-linked structure plays a decisive role in determining the mechanical properties of polymer materials.Although various supramolecular polymer networks(SPNs)based on noncovalent bonds have been developed,few s...The cross-linked structure plays a decisive role in determining the mechanical properties of polymer materials.Although various supramolecular polymer networks(SPNs)based on noncovalent bonds have been developed,few studies have focused on constructing SPNs through dual host-guest interactions between macro-and small-molecular building blocks.Herein,we utilized these building blocks to prepare SPNs aimed at addressing this gap.Specifically,the supramolecular polymer network SPN-EF was prepared through crown ether and pillararenebased dual host-guest interactions.Compared to the control systems SPN-AC and SPN-BD,which incorporate only single type of host-guest interaction(based on either crown ether or pillararene),the SPN-EF integrate the advantages of both systems,exhibiting balanced and enhanced mechanical properties.Moreover,the resulting SPN gels retain significant dynamic properties,including excellent self-healing capability and stimuliresponsiveness.展开更多
Self-sustaining microalgal-bacterial membrane photobioreactors(SSMB-MPBRs)have attracted growing attention due to energy conversation and carbon reduction potential.However,microalgae proliferation,which is essential ...Self-sustaining microalgal-bacterial membrane photobioreactors(SSMB-MPBRs)have attracted growing attention due to energy conversation and carbon reduction potential.However,microalgae proliferation,which is essential for pollutant removal,exacerbates membrane fouling,and the underly mechanisms remain unclear.In this study,pollutant removal efficacy and membrane fouling in a SSMB-MPBR process were investigated at different inoculation microalgae to bacteria(M/B)ratios.The intrinsic mechanisms were examined based on algae-bacteria cooperative and competitive interactions.The results showed that pollutant removal efficacy and chlorophyll-a productivity increased,and decreased with the inoculation M/B ratio from 1:1 to 3:1,in line with the rate of algae-bacteria interaction.The highest pollutant removal efficiency(93.65% of chemical oxygen demand(COD),85.42% of total nitrogen(TN),91.65%of total phosphorus(TP))and chlorophyll-αproductivity(0.82 mg/(L·day))were achieved at inoculation M/B ratio of 2:1 due to the selective enrichment of microalgae and their competitions with predominant functional bacteria,which increased the upregulation of functional enzymes and relative metabolism.Similar trends were observed for the fouling behavior,which corresponded to the algae-bacteria competitive interaction.The increased involution stress promoted biosynthesis and hydrophobicity of extracellular polymeric substances(EPS),which,combined with small free microalgae,contributed to a compact bio-cake,and the most significant membrane fouling.These findings provide mechanistic insights into pollutant removal and membrane fouling and highlight the need to optimize algae-bacteria cooperative and competitive interactions to increase the efficiency and sustainability of the SSMB-MPBR process in treating municipal wastewater.展开更多
Porous hydrogel sensors have attracted significant attention in fields such as smart wearables and medical monitoring due to their high sensitivity.However,existing fabrication methods typically degrade the surface sm...Porous hydrogel sensors have attracted significant attention in fields such as smart wearables and medical monitoring due to their high sensitivity.However,existing fabrication methods typically degrade the surface smoothness of hydrogels when introducing porous structures and face significant challenges in removing fillers completely.To address these challenges,we herein introduce a novel one-step,thermosensitive spray-coating technique for the preparation of aircell hydrogel(ACH).This method leverages the rapid cooling of a thermoresponsive gelatin methacryloyl solution through atomization,enabling rapid cross-linking within seconds and air bubbles encapsulated in situ.Additionally,the transient flow of the pre-gel facilitates the repair of voids formed by ruptured surface bubbles,leading to the creation of the ACH with uniformly distributed inner air bubbles and a smooth outer surface.The mold-free fabrication method is independent of substrate surface properties,enabling the creation of a porous hydrogel film with a thickness as thin as 163 µm.Furthermore,the dual-crosslinked network endows the ACH with excellent anti-swelling properties,and the physical crosslinking between gelatin molecules allows the ACH to self-heal.The ACH exhibits excellent sensitivity in deformation sensing and can even successfully track minor external forces,which enables it to effectively complete various tasks such as facial expression recognition,pitch differentiation,and motion detection.By integrating the ACH into a sensing glove,we also demonstrate the significant potential of the ACH for applications in human-machine interaction and tactile sensing.Ultimately,the ACH sensors are also applied to motion mapping and machine tactile feedback,indicating their promising potential in human-machine interaction.展开更多
An integrated real-time control methodology is introduced to mitigate process-induced challenges,namely temperature overshoot,uneven cure,and interfacial shear stress,during the autoclave curing of Carbon Fiber-Reinfo...An integrated real-time control methodology is introduced to mitigate process-induced challenges,namely temperature overshoot,uneven cure,and interfacial shear stress,during the autoclave curing of Carbon Fiber-Reinforced Polymer(CFRP)composites.First,a high-fidelity Finite Element(FE)model incorporating tool-part interaction is developed to reveal the curing process of the composites,wherein the interaction is characterized by friction interface modeling with experimentally measured cure-dependent friction coefficients.The accuracy of FE model is confirmed through experimental tests on a doubly curved T-stiffened composite panel.This validated model then generates a dataset of curing temperature profile and associated defect information,which is used to train a customized Long Short-Term Memory(LSTM)neural network.We culminate in a real-time control framework that actively optimizes the curing process by integrating LSTM-based state prediction with Q-learning-driven decision logic.The optimized thermal profile demonstrates a clear performance enhancement over the traditional multi-dwell approach,achieving marked reductions in temperature difference,Degree of Cure(DoC)difference and tool-part interface shear stress,which provides more insights for intelligent composite manufacturing.展开更多
基金funded by the U.S.Department of Education under Grant Number ED#P116S210005the National Science Foundation under Grant Numbers 2226936 and 2420405.
摘要Theintegration of human factors into artificial intelligence(AI)systems has emerged as a critical research frontier,particularly in reinforcement learning(RL),where human-AI interaction(HAII)presents both opportunities and challenges.As RL continues to demonstrate remarkable success in model-free and partially observable environments,its real-world deployment increasingly requires effective collaboration with human operators and stakeholders.This article systematically examines HAII techniques in RL through both theoretical analysis and practical case studies.We establish a conceptual framework built upon three fundamental pillars of effective human-AI collaboration:computational trust modeling,system usability,and decision understandability.Our comprehensive review organizes HAII methods into five key categories:(1)learning from human feedback,including various shaping approaches;(2)learning from human demonstration through inverse RL and imitation learning;(3)shared autonomy architectures for dynamic control allocation;(4)human-in-the-loop querying strategies for active learning;and(5)explainable RL techniques for interpretable policy generation.Recent state-of-the-art works are critically reviewed,with particular emphasis on advances incorporating large language models in human-AI interaction research.To illustrate some concepts,we present three detailed case studies:an empirical trust model for farmers adopting AI-driven agricultural management systems,the implementation of ethical constraints in roboticmotion planning through human-guided RL,and an experimental investigation of human trust dynamics using a multi-armed bandit paradigm.These applications demonstrate how HAII principles can enhance RL systems’practical utility while bridging the gap between theoretical RL and real-world human-centered applications,ultimately contributing to more deployable and socially beneficial intelligent systems.
摘要数智时代,用户与人工智能系统的交互模式正在发生怎样的转变?《Human-AI Interaction and Collaboration》一书通过理论框架构建、应用场景描述、潜在风险辨识,为读者全方位地解析了人智交互与协作的内在逻辑,提出“以人为本”这一人工智能系统设计的首要原则,详细分析了用户感知、系统设计、人智关系等一系列因素如何影响人智协作的效率与成果,在针对潜在风险制定应对方案的同时,为读者展示了人智协作在医疗、科研、金融等多学科场景下的广阔前景。该书为未来人智交互与协作相关研究提供了坚实的理论框架,也为人工智能系统设计者提供了伦理指南。
摘要Interactional naturalness of human-AI dialogue is a key factor in fostering meaningful communicative engagement.This study addresses this gap by adopting Conversation Analysis(CA)as a methodological framework to examine recorded speaking practice sessions between English as a Second or Foreign Language(ESL/EFL)learners and an AI tutoring system.Drawing on the principles of emergence,indexicality,and recipient design,the study investigates why and how interactional naturalness is violated in AI-mediated conversation.These findings reveal that while the AI tutor produces grammatically well-formed and topically relevant utterances,its conduct frequently diverges from the moment-by-moment,context-sensitive organization that characterizes natural human interaction.Particularly,the AI system fails to consistently align with emergent turn sequences,indexical references,and recipient-designed responses.The results indicate that current AI tutoring systems,despite advanced language models,have not yet achieved interactional naturalness as defined from a conversation-analytic perspective.Therefore,a shift in evaluation criteria from technical accuracy to interactional quality in the design and assessment of AI-based language tutors is called-for.
基金supported by the National Natural Science Foundation of China(No.12372233)the Innovation Foundation for Doctor Dissertation of Northwestern Polytechnical University,China(No.25GH01020005)the“111 Project”of China(No.B17037)。
摘要As a multidisciplinary phenomenon,panel aeroelasticity in shock-dominated flow is featured by two primary interactions:Fluid-Structure Interactions(FSIs)and Shock-Boundary Layer Interactions(SBLIs).The former raises structural concerns,and the latter is of aerodynamic interest.Thus,panel aeroelasticity in shock-dominated flow represents a vital topic for the development and optimization of supersonic vehicles and propulsion systems.This review systematically summarizes recent advances in the methodologies applied to capture structural and fluid dynamics,including theoretical models,numerical simulations,and wind tunnel experiments.The application of data-driven modal decomposition,an advanced technique to extract physically crucial features,on the topic is introduced.From the perspective of FSIs,the distinctive aeroelastic behaviors in shock-dominated flow,including hysteresis phenomena and nonlinear responses,are highlighted.From the perspective of SBLIs,the modifications in their spatial and temporal characteristics imposed by the aeroelastic responses are emphasized.Motivated by the interaction between the shock waves and structural response,different strategies have been proposed to implement aeroelastic suppression and shock control,which have the potential to enhance structural safety and aerodynamic performance in the next generation of high-speed flight vehicles.
基金supported,in part,by the National Science Foundation under grants CHE 2404209 and CHE 2320239.
摘要The characterization of drug-target interactions is a key component of drug discovery,testing,and development.Affinity chromatography is one approach that can be used for this type of analysis.For instance,this may be done by using an immobilized target as a stationary phase and a drug as the applied solute.This review will discuss the various ways in which affinity chromatographic methods have been used to examine drug-target interactions,with an emphasis on high-performance methods.The general principles of this approach and factors to consider in its use for drug-target interaction analysis will first be examined.Methods based on zonal elution or frontal analysis for binding and competition studies will then be discussed.Various techniques for kinetic studies will next be considered,along with approaches that employ secondary binding agents and hybrid techniques.In each case,the general principles and theory of an approach will be given along with examples of its use in drug-target interaction studies.Advantages or limitations of each approach will be provided as well.This information should make it possible in the future to extend these techniques to other drug-target systems of interest in biomedical research and drug testing or development.
基金Support of the State of Rio de Janeiro(FAPERJ),Brazil(Grant Nos.:E-26/210.017/2024,E-200.172/2023,E-26/200.165/2024,E-26/200.164/2024,and E-26/210.547/2025)the Coordination for the Improvement of Higher Education Personnel(CAPES),Brazil(Finance Code 001)National Council for Scientific and Technological Development(CNPq),Brazil(Grant Nos.:307108/2021-0 and 302464/2022-0)for their support.
摘要Affinity selection mass spectrometry(AS-MS)has emerged as a powerful label-free technique for identifying and characterizing ligand-target interactions.This review explores the diverse applications of AS-MS in drug discovery,including its role in selective screening,binding site characterization,and quantitative affinity determination.We discuss the use of AS-MS for determining equilibrium dissociation constants(KD)and competitive binding parameters(affinity competition experiment 50%(ACE50)),highlighting its ability to rank ligand affinities efficiently.The review also examines AS-MS applications in fragment-based drug discovery(FBDD),screening for molecular glues,and investigating interactions with membrane proteins.Moreover,we address key technical challenges,including competitive binding effects,protein stability,and ligand dissociation kinetics,along with recent advancements in automation and artificial intelligence(AI)integration.Rather than providing a comprehensive literature review,this work aims to broaden the applicability of AS-MS assays and encourage researchers to explore its use in underutilized contexts.By providing rapid and high-sensitivity affinity measurements,AS-MS continues to expand its role in drug discovery and structural biology,complementing conventional biophysical techniques.
基金financially supported by the National Natural Science Foundation of China(Grant No.52171259)the National Key Technologies Research and Development Program(Grant No.2022YFE0107000)+1 种基金the High-tech Ship Research Project of the Ministry of Industry and Information Technology(Grant No.[2021]342)the Young Scientists Fund of the National Natural Science Foundation of China(Grant No.52301331).
摘要Structure-ice interaction problems have attracted increasing attention,yet accurately predicting the loads exerted by sea ice on ship hulls remains a significant challenge.Over the past few decades,various numerical methods have been employed to simulate ice resistance on ships and evaluate their manoeuvrability during ice-structure interaction.Among these approaches,the circumferential crack method has demonstrated both high efficiency and accuracy.This paper provides a detailed introduction to the fundamental theory of this method,including the numerical modeling of different failure modes and the dynamic ice motion responses.Furthermore,it reviews existing studies on predicting ice resistance and assessing the manoeuvrability of icebreakers navigating through ice-covered regions using the circumferential crack method.Several recommendations for future research in this field are also presented.
基金supported by Teaching Development Grant from the University of Hong Kong,China.
摘要Research in human‑robot Interaction(HRI)has increasingly demonstrated how Augmented Reality(AR)enables better interactions between humans and robots.However,the design of HRI remains less understood.Through a systematic literature review of 53 related papers,this research provides an overview of the emerging applications and trends for AR and identifies three types of AR interfaces as follows:1)remote modular interface,2)proximal modular interface,and 3)proximal integral interface.The review indicates potential future directions of construction‑oriented and human‑centric interaction design studies,leading to four pairs of subsystems,which are frequently modularised or integrated,and three conceptual frameworks for HRI interfaces are proposed.Moreover,this research contributes to the theoretical exploration of interaction design.Future applications can adapt to various tasks by using the proposed three conceptual frameworks for interfaces,as well as combining the four proposed subsystem pairs to suit specific task requirements in the construction sector.
基金support from the National Natural Science Foundation of China(No.22308279)Guangdong Basic and Applied Basic Research Foundation(No.2021A1515110695)Natural Science Foundation of Chongqing(No.2023NSCQMSX2773).
摘要As a common electronic adhesive,ultraviolet(UV)curing polyurethane acrylate adhesive has both flexibility and wear resistance of polyurethane,excellent weather resistance and optical properties of acrylate.Despite the extensive applications,it is still difficult to solve the problems caused by the shrinkage of adhesive.Here,a new type of photosensitive adhesive for bonding electronic components based on supramolecular interaction was designed and synthesized.The supramolecular interaction of cyclodextrin and adamantane moieties introduced into the adhesive polymer entitles the viscosity of the adhesive to rise rapidly during use,thereby preventing adhesive loss and dislocation of electronic components.UV light could further cure the adhesive and position the electronic components.The adhesive shrunk<2%when cured by UV light,so it can be used for electronic packaging and high-resolution,defect-free lithography.
基金supported by the National Natural Science Foundation of China(32072205)the Science and Technology Plan Project of Shaoxing City(2022B43001)。
摘要This study aimed to gain insight into the interaction and metabolic mechanism between Aspergillus flavus U-16 and Saccharomyces cerevisiae HJ during co-fermentation.Despite the suppressed transcriptional activity of SU-16,its metabolites exerted a pivotal regulatory influence.The metabolites produced by SU-16 significantly increased the ethanol content to 17.0%.They also significantly increased the contents of crucial flavor substances,including amino acids(at least 5 times)and esters(up to 1100 mg/L).SU-16 produced 130 metabolites that may affect flavor compounds,while HJ mainly controlled the synthesis of 161 compounds.Metabolites of SU-16 with different molecular weights were observed to exert differential regulatory effects on the fermentation process and the metabolic pathway of HJ.The small molecule metabolites of SU-16 promoted the growth of HJ significantly.In contrast,its macromolecular metabolites were found to mainly affect the expression of genes related to carbohydrate and amino acid metabolism in the HJ genome.
基金supported by the China Postdoctoral Science Foundation(2025M772118)the National Natural Science Foundation of China(82271525,82471553,and 82471507)the Natural Science Foundation of Guangdong Province(2023A1515010456 and 2024A1515030100).
摘要Schizophrenia(SCZ)is a severe mental illness influenced by gene-environment interactions(GEI).However,little is known about how GEI mediates SCZ.The present study aimed to examine the underlying mechanism of SCZ mediated by GEI.We found that a single environmental factor(two-week adolescent social isolation)or genetic factor(the heterozygous Rims1 knockout mice)did not induce SCZ-like behaviors.Interestingly,two-week adolescent social isolation successfully caused SCZ-like behaviors in heterozygous Rims1 knockout mice,which can be rescued by anti-SCZ drugs.RNA-seq analysis further revealed that synaptic vesicle-related biological processes and target genes were enriched in the hippocampus of GEI animal model mice,which was accompanied by disturbed excitatory synaptic neurotransmission.Finally,the Nrg1 gene was decreased in our RNA-seq analysis,and supplementation of Nrg1 ameliorated SCZ-like behaviors in heterozygous Rims1 socially isolated mice.Our findings establish a novel GEI animal model and offer a potential therapeutic target in the treatment of SCZ.
摘要Objectives:Pancreatic ductal adenocarcinoma(PDAC)is a leading cause of cancer-related mortality and mainly affects older adults,who frequently experience polypharmacy.While systemic therapy may improve outcomes in selected older patients,the use of multiple drugs increases the risk of potential drug–drug interactions(pDDIs).This study aimed to evaluate the prevalence and characteristics of pDDIs in older patients with PDAC receiving first-line systemic therapy and their potential impact on clinical outcomes.Methods:We conducted a retrospective single-center study including patients aged≥75 years with PDAC who initiated first-line systemic therapy between December 2011 and January 2023.Prescription data were retrieved from institutional and outpatient pharmacy databases,and all concomitant medications were recorded.pDDIs were assessed using Intercheck Web and classified as A(minor),B(moderate),C(major but manageable),or D(contraindicated).Patients(n=140)were stratified according to the number of active pharmaceutical ingredients(APIs)at treatment initiation:20(n=44).Results:A higher number of APIs was significantly associated with an increased burden of pDDIs,including A,B,C,and D interactions(all p20 APIs showing the highest prevalence of clinically relevant interactions.Overall,70%of patients had at least one pDDI,but no significant differences were observed in 1-year overall survival,access to second-line therapy,or chemotherapy type across API groups.Conclusion:Polypharmacy significantly increases the risk of pDDIs in older patients with PDAC,but this did not translate into differences in 1-year survival or access to subsequent therapy.
基金supported by the National Natural Science Foundation of China(Nos.22577103,22571257,22301246)the Project of Science and Technology of Social Development in Shaanxi Province(Nos.2023-YBSF-151,2024SF-YBXM-294)+2 种基金the Shaanxi Province Postdoctoral Science Foundation(No.2023BSHEDZZ101)the Fundamental Research Program of Shanxi Province(No.202403021221212)the Research Foundation Project of Changzhi Medical College(No.HZZD202406)。
摘要Biomolecular self-assembly systems form the cornerstone of biological structures,in which the interactions between carbohydrates and proteins play a crucial role in many life processes.This has driven the development of biomimetic carbohydrate-protein supramolecular assemblies(BCPSAs),which show great potential in biomedical research.We successfully constructed a novel supramolecular hybrid nanocarrier with pH responsiveness and targeting capabilities based on specific interactions between carbohydrates and proteins.This system self-assembles into nanoparticles CA@MP5 in water through the specific binding of mannose clusters on glycosylated pillar[5]arene MP5 with lectin Con A.Additionally,through a host-guest assembly strategy with the synthesized β-D-galactopyranosyl pyridine derivative G,it complexes within the MP5 cavity to form a"double-sweet"supramolecular glyco-nanoprotein(CA@MP5⊃G).This system shows high affinity for the asialoglycoprotein receptor(ASGPR)on HepG2 liver cancer cells.In vitro,the doxorubicin(DOX)-loaded micelles(DCA@MP5⊃G)release drug effectively in acidic environments and significantly inhibit HepG2 cell growth,with reduced toxicity toward normal cells.In vivo,the nanocarrier targets tumors effectively,reducing systemic toxicity and inhibiting tumor growth.This study offers a novel construction strategy for drug delivery systems based on carbohydrate-protein interactions in anti-tumor applications,which is significant for enhancing therapeutic efficacy.
基金supported by the NHMRC Healthy Environments and Lives(HEAL)National Research Network and Scheme for Promotion of Academic and Research Collaboration(SPARC)(No.SPARC/2019-2020/P1789/SL)。
摘要Airborne microplastics(MPs)are prevalent indoors,and due to their low aerodynamic diameter,they can be inhaled,posing potential health risks.Although the toxic effects of airborne MPs have been explored using in vivo and in vitro models,the interactions between MPs and cellular receptors remain understudied.In this research,dust samples from a confined space within an academic setting were investigated for microplastics(MPs)and their abundance.Further,the most prevalent microplastics were used to study receptor binding and competition interaction studies using the In-Silico method against natural agonists and antagonists of major inflammatory receptors,including the human platelet-activating factor receptor(PAFR),C-X-C motif chemokine receptor 1(CXCR1),β2-adrenergic receptor(β2-AR),and toll-like receptor 2(TLR-2).Results revealed that polyester(PE)was the predominant polymer,accounting for 23.96%of the samples.Analysis indicated that a monomer of PET,Ethylene Terephthalate(ET),exhibited a high binding affinity of-6.5 kcal/mol with the PAFR receptor.Additionally,the molecular dynamics and protein-ligand interaction study,which involves hydrogen bonding,explains the differential binding effect of ET and the control compound with the targeted receptor.The complex was formed between the ET and the receptors ofβ2-AR,CXCR1,and TLR-2,with a maximum of seven and a minimum of one hydrogen bond throughout the simulation.This research lays a foundation for understanding the potential health implications of MPs in confined office spaces,underscoring the need for further in vivo and in vitro examinations.
基金supported by the National Key Research and Development Program of China(Grant No.2024YFC3013400)the National Natural Science Foundation of China(Grant No.42471031).
摘要Amid accelerating climate change and intensifying human activities,flood risks have been accumulating,rein forcing the bidirectional interplay between humans and floods.Within this context,human-flood interaction has emerged as a rapidly expanding research frontier that is attracting growing attention.Here,we review 78 rep resentative studies to examine publication trends,research hotspot shifts,and persistent challenges in this field,combining bibliometric analyses with flood disaster records.We found that,over the past three decades,publica tions on human-flood interaction have increased exponentially,with their focus shifting from one-way impacts towards two-way human-flood coupling.The global collaboration network displays a“dual-core”structure,with transcontinental collaboration between China and the United States and regional collaborations within Europe,whereas participation from Global South countries remains marginal.In addition,a spatial disconnect persists between the geographic distribution of research hotspot regions and that of flood records.Based on bibliometric mapping,we further highlight three enduring challenges:(i)scale mismatches,(ii)spatiotemporal coupling gaps,and(iii)deficits in behavioral modeling.The insights gained from this study will help guide future research and investment in human-flood interaction,with practical implications for enhancing resilience to flood disasters and promoting sustainable human-water coexistence.
基金supported by the PolyU Postdoc Matching Fund Scheme(1-W327)the PolyU Grant(1-CE0H)+4 种基金the RGC SRFS 2122-5S02the Natural Science Foundation of Guangdong Province(2024A1515010639)the Shenzhen Key Laboratory for Low-Carbon Construction Material and Technology(ZDSYS20220606100406016)the National Key Laboratory of Green and Long-Life Road Engineering in Extreme Environment(Shenzhen)(868-000003010103)the Scientific Foundation for Youth Scholars of Shenzhen University(806-0000340606)。
摘要Hydrogels,with their hydrophilicity,flexibility,and environmental friendliness,are highly desirable for moisture-electric generators(MEGs)that harness ubiquitous moisture to generate electrical energy.As the active material layer in MEGs,hydrogels play a crucial role in absorbing atmospheric moisture and converting chemical potential energy into electricity.However,the relatively low output current of the device and the instability of hydrogels pose challenges to the development of high-performance hydrogel-based MEGs.Herein,we introduce a straightforward,feasible,cost-effective,and versatile two-step solvent displacement strategy to overcome the barrier associated with the development of MEGs.Through tunable solvent interactions of glycerol and water,the moisture absorption capability and stability of the hydrogel can be improved,while promoting favorable ion migration.Such an effective processing route not only significantly boosts the output performances but also greatly improves the long-term durability of hydrogel-based MEGs.Notably,the current output and power density of the treated MEGs can increase by up to two orders of magnitude.The mechanisms behind the intriguing observation are investigated by various characterizations and theoretical calculations.This universal strategy holds promise to be extended to various hydrogel-based MEGs.Moreover,the MEGs can be used for energy harvesting,self-powered respiratory monitoring,and non-contact humidity detection.This work offers new opportunities for advancing green energy and self-powered technologies.
基金financially supported by the National Natural Science Foundation of China(Nos.22561023,22525107 and 22161020)the Natural Science Foundation of Jiangxi Province(No.20252BAC240279)。
摘要The cross-linked structure plays a decisive role in determining the mechanical properties of polymer materials.Although various supramolecular polymer networks(SPNs)based on noncovalent bonds have been developed,few studies have focused on constructing SPNs through dual host-guest interactions between macro-and small-molecular building blocks.Herein,we utilized these building blocks to prepare SPNs aimed at addressing this gap.Specifically,the supramolecular polymer network SPN-EF was prepared through crown ether and pillararenebased dual host-guest interactions.Compared to the control systems SPN-AC and SPN-BD,which incorporate only single type of host-guest interaction(based on either crown ether or pillararene),the SPN-EF integrate the advantages of both systems,exhibiting balanced and enhanced mechanical properties.Moreover,the resulting SPN gels retain significant dynamic properties,including excellent self-healing capability and stimuliresponsiveness.
基金supported by Hebei Natural Science Foundation(No.E2025209010)the Scientific Research Foundation of North China University of Science and Technology。
摘要Self-sustaining microalgal-bacterial membrane photobioreactors(SSMB-MPBRs)have attracted growing attention due to energy conversation and carbon reduction potential.However,microalgae proliferation,which is essential for pollutant removal,exacerbates membrane fouling,and the underly mechanisms remain unclear.In this study,pollutant removal efficacy and membrane fouling in a SSMB-MPBR process were investigated at different inoculation microalgae to bacteria(M/B)ratios.The intrinsic mechanisms were examined based on algae-bacteria cooperative and competitive interactions.The results showed that pollutant removal efficacy and chlorophyll-a productivity increased,and decreased with the inoculation M/B ratio from 1:1 to 3:1,in line with the rate of algae-bacteria interaction.The highest pollutant removal efficiency(93.65% of chemical oxygen demand(COD),85.42% of total nitrogen(TN),91.65%of total phosphorus(TP))and chlorophyll-αproductivity(0.82 mg/(L·day))were achieved at inoculation M/B ratio of 2:1 due to the selective enrichment of microalgae and their competitions with predominant functional bacteria,which increased the upregulation of functional enzymes and relative metabolism.Similar trends were observed for the fouling behavior,which corresponded to the algae-bacteria competitive interaction.The increased involution stress promoted biosynthesis and hydrophobicity of extracellular polymeric substances(EPS),which,combined with small free microalgae,contributed to a compact bio-cake,and the most significant membrane fouling.These findings provide mechanistic insights into pollutant removal and membrane fouling and highlight the need to optimize algae-bacteria cooperative and competitive interactions to increase the efficiency and sustainability of the SSMB-MPBR process in treating municipal wastewater.
基金financially supported by the National Key R&D Program of China(Grant No.2023YFE0108900)EU HORIZON 2021 L4DNANO(No.101086227)。
摘要Porous hydrogel sensors have attracted significant attention in fields such as smart wearables and medical monitoring due to their high sensitivity.However,existing fabrication methods typically degrade the surface smoothness of hydrogels when introducing porous structures and face significant challenges in removing fillers completely.To address these challenges,we herein introduce a novel one-step,thermosensitive spray-coating technique for the preparation of aircell hydrogel(ACH).This method leverages the rapid cooling of a thermoresponsive gelatin methacryloyl solution through atomization,enabling rapid cross-linking within seconds and air bubbles encapsulated in situ.Additionally,the transient flow of the pre-gel facilitates the repair of voids formed by ruptured surface bubbles,leading to the creation of the ACH with uniformly distributed inner air bubbles and a smooth outer surface.The mold-free fabrication method is independent of substrate surface properties,enabling the creation of a porous hydrogel film with a thickness as thin as 163 µm.Furthermore,the dual-crosslinked network endows the ACH with excellent anti-swelling properties,and the physical crosslinking between gelatin molecules allows the ACH to self-heal.The ACH exhibits excellent sensitivity in deformation sensing and can even successfully track minor external forces,which enables it to effectively complete various tasks such as facial expression recognition,pitch differentiation,and motion detection.By integrating the ACH into a sensing glove,we also demonstrate the significant potential of the ACH for applications in human-machine interaction and tactile sensing.Ultimately,the ACH sensors are also applied to motion mapping and machine tactile feedback,indicating their promising potential in human-machine interaction.
基金provided by the National Key Research and Development Program of China(No.2021YFB3401700)the National Natural Science Foundation of China(Nos.12302189 and 12220101002)the Shaanxi Provincial Key Science and Technology Innovation Team,China(No.2023-CX-TD-14)。
摘要An integrated real-time control methodology is introduced to mitigate process-induced challenges,namely temperature overshoot,uneven cure,and interfacial shear stress,during the autoclave curing of Carbon Fiber-Reinforced Polymer(CFRP)composites.First,a high-fidelity Finite Element(FE)model incorporating tool-part interaction is developed to reveal the curing process of the composites,wherein the interaction is characterized by friction interface modeling with experimentally measured cure-dependent friction coefficients.The accuracy of FE model is confirmed through experimental tests on a doubly curved T-stiffened composite panel.This validated model then generates a dataset of curing temperature profile and associated defect information,which is used to train a customized Long Short-Term Memory(LSTM)neural network.We culminate in a real-time control framework that actively optimizes the curing process by integrating LSTM-based state prediction with Q-learning-driven decision logic.The optimized thermal profile demonstrates a clear performance enhancement over the traditional multi-dwell approach,achieving marked reductions in temperature difference,Degree of Cure(DoC)difference and tool-part interface shear stress,which provides more insights for intelligent composite manufacturing.