Unlike mammals,zebrafish possess a remarkable ability to regenerate their spinal cord after injury,making them an ideal vertebrate model for studying regeneration.While previous research has identified key cell types ...Unlike mammals,zebrafish possess a remarkable ability to regenerate their spinal cord after injury,making them an ideal vertebrate model for studying regeneration.While previous research has identified key cell types involved in this process,the underlying molecular and cellular mechanisms remain largely unexplored.In this study,we used single-cell RNA sequencing to profile distinct cell populations at different stages of spinal cord injury in zebrafish.Our analysis revealed that multiple subpopulations of neurons showed persistent activation of genes associated with axonal regeneration post injury,while molecular signals promoting growth cone collapse were inhibited.Radial glial cells exhibited significant proliferation and differentiation potential post injury,indicating their intrinsic roles in promoting neurogenesis and axonal regeneration,respectively.Additionally,we found that inflammatory factors rapidly decreased in the early stages following spinal cord injury,creating a microenvironment permissive for tissue repair and regeneration.Furthermore,oligodendrocytes lost maturity markers while exhibiting increased proliferation following injury.These findings demonstrated that the rapid and orderly regulation of inflammation,as well as the efficient proliferation and redifferentiation of new neurons and glial cells,enabled zebrafish to reconstruct the spinal cord.This research provides new insights into the cellular transitions and molecular programs that drive spinal cord regeneration,offering promising avenues for future research and therapeutic strategies.展开更多
The development of advanced support is conducive to promoting the practical application of fuel cells but remains an enormous challenge in terms of stabilizing catalyst particles and enabling improved accessibility to...The development of advanced support is conducive to promoting the practical application of fuel cells but remains an enormous challenge in terms of stabilizing catalyst particles and enabling improved accessibility to O_(2).Beyond solid carbon black and conventional porous carbon,we demonstrate a new type of mesoporous bowl-like carbon support with a high specific surface area of over 1200 m^(2)g^(-1)and~4 nm pore.Both rotating disk electrode and membrane electrode assembly(MEA)tests show that BC-supported Pt_(3)Co(Pt_(3)Co/BC)catalyst greatly outperforms hollow porous carbon spheres and solid carbon spheres supported Pt_(3)Co catalysts.The Pt_(3)Co/BC catalysts exhibited remarkable performance as cathode catalyst in MEA,achieving a comparable open-circuit voltage of approximately 0.95 V under H_(2)-air condition and a current density of 1.3 A cm^(-2)at 0.6 V with a loading of only 0.2 mgPt cm^(-2).Diffusion simulations and physical characterizations demonstrate that the high porosity and highly accessible pore structure of BC support facilitate the uniform distribution of catalyst particles and enhance the mass transport of O_(2),thereby resulting in a significant improvement in catalytic activity and durability.This work provides new insights into the influence of support shape on mass transport of reactants and electrochemical performances of the catalyst in MEA.展开更多
The development of efficient and cost‐effective metal‐free electrocatalysts for oxygen reduction reaction(ORR)has become crucial for electrochemical energy systems.However,reasonably validating and precisely regulat...The development of efficient and cost‐effective metal‐free electrocatalysts for oxygen reduction reaction(ORR)has become crucial for electrochemical energy systems.However,reasonably validating and precisely regulating the active sites for designing optimized materials are still challenging.Herein,we report a precise and controllable tandem strategy to boost the ORR activity based on metal‐free covalent organic frameworks(MFCOFs)comprising imine‐N,thiophene‐S,or triazine‐N.Among these MFCOFs,post‐tandem BTT‐TAT‐COF structure displayed a more positive catalytic capability and excellent electrochemical stability,indicating that the synergistic catalysis of multiple active sites induced the ORR catalytic activity through the conjugated skeleton of the structure.Density‐functional theory calculations suggest that the series‐connected backbone contained highly effective electrocatalytic active centers and provided synergistic catalysis.More importantly,this strategy highlights new opportunities for the advancement of efficient COF‐based metal‐free ORR catalysts.展开更多
The electronic conductivity and durability of porous carbon supports can be improved by increasing the degree of graphitization in the material;however,the preparation of highly graphitic porous carbon using conventio...The electronic conductivity and durability of porous carbon supports can be improved by increasing the degree of graphitization in the material;however,the preparation of highly graphitic porous carbon using conventional furnaces remains a significant challenge.Herein,we demonstrate a universal and highly efficient carbothermal shock strategy that significantly improves the degree of graphitization of porous carbon supports,including bowl‐like carbon,hollow carbon spheres,ZIF8‐derived carbon,BP2000,and Ketjen EC 300J.Taking bowl‐like carbon as an example,we illustrate the synthesis of a graphitized bowl‐like carbon(G‐BC‐S)support and evaluate the performance of PtCo/G‐BC‐S in the oxygen reduction reaction(ORR)in rotating disk electrodes(RDE)and H_(2)/air PEM single cells.PtCo/G‐BC‐S exhibits faster ORR kinetics than PtCo/BC and Pt/C,with little loss of activity(25%)and only 13 mV of E1/2 decay after 20000 cycles accelerated stress testing under 1.0-1.5 V vs.a reversible hydrogen electrode(RHE).The significantly enhanced performance of the PtCo/G‐BC‐S catalyst arises from the high activity and chemical/structural stability of the PtCo intermetallic nanoparticles and from the high degree of graphitization and well‐defined porous structure of the bowl‐like carbon support,which confers excellent electrical conductivity and oxygen transport properties.This study provides a reliable and universal strategy for the development of high‐performance porous carbon supports for practical applications in fuel cells.展开更多
Interpreting deep neural networks is of great importance to understand and verify deep models for natural language processing(NLP)tasks.However,most existing approaches only focus on improving the performance of model...Interpreting deep neural networks is of great importance to understand and verify deep models for natural language processing(NLP)tasks.However,most existing approaches only focus on improving the performance of models but ignore their interpretability.In this work,we propose a Randomly Wired Graph Neural Network(RWGNN)by using graph to model the structure of Neural Network,which could solve two major problems(word-boundary ambiguity and polysemy)of ChineseNER.Besides,we develop a pipeline to explain the RWGNNby using Saliency Map and Adversarial Attacks.Experimental results demonstrate that our approach can identify meaningful and reasonable interpretations for hidden states of RWGNN.展开更多
Spinal cord injury(SCI),a complex neurological disorder,triggers a series of devastating neuropathological events such as ischemia,oxidative stress,inflammatory events,neuronal apoptosis,and motor dysfunction.However,...Spinal cord injury(SCI),a complex neurological disorder,triggers a series of devastating neuropathological events such as ischemia,oxidative stress,inflammatory events,neuronal apoptosis,and motor dysfunction.However,the classical necrosome,which consists of receptor-interacting protein(RIP)1,RIP3,and mixed-lineage kinase domain-like protein,is believed to control a novel type of programmed cell death called necroptosis,through tumour necrosis factor-alpha/tumour necrosis factor receptor-1 signalling or other stimuli.Several studies reported that necroptosis plays an important role in neural cell damage,release of intracellular pro-inflammatory factors,lysosomal dysfunction and endoplasmic reticulum stress.Recent research indicates that necroptosis is crucial to the pathophysiology of a number of neurological disorders and SCIs.In our review,we summarize the potential role of programmed cell death regulated by necroptosis in SCI based on its molecular and pathophysiological mechanisms.We also summarize the targets of several necroptosis pathways,which provide a more reliable reference for the treatment of SCI.展开更多
The pattern of thematic progression,reflecting the semantic relationships between contextual two sentences,is an important subject in discourse analysis.We introduce a new corpus of Chinese news discourses annotated w...The pattern of thematic progression,reflecting the semantic relationships between contextual two sentences,is an important subject in discourse analysis.We introduce a new corpus of Chinese news discourses annotated with thematic progression information and explore some computational methods to automatically extracting the discourse structural features of simplified thematic progression pattern(STPP)between contextual sentences in a text.Furthermore,these features are used in a hybrid approach to a major discourse analysis task,Chinese coreference resolution.This novel approach is built up via heuristic sieves and a machine learning method that comprehensively utilizes both the top-down STPP features and the bottom-up semantic features.Experimental results on the intersection of the CoNLL-2012 task shared dataset and the CDTC corpus demonstrate the effectiveness of our proposed approach.展开更多
The differences in Zn crystal plane kinetics can lead to non-uniform deposition,promoting dendrite growth and side reactions,especially under high deposition capacities.Fast kinetics can also cause anion depletion on ...The differences in Zn crystal plane kinetics can lead to non-uniform deposition,promoting dendrite growth and side reactions,especially under high deposition capacities.Fast kinetics can also cause anion depletion on the zinc anode surface,leading to uneven electric field distribution and worsening these issues.Inducing preferred electrodeposition of the Zn(101)crystal plane can ensure dense epitaxial growth and achieve fast reaction kinetics.However,its highly reactive and wave-like arrangement will also lead to higher hydrogen evolution activity and cause uneven electric field distribution,accelerating side reactions and dendrite growth.This study utilizes the adsorption ability of 2-mercaptoethanesulfonate(MES)anion on the zinc anode surface to optimize the interfacial concentration and electric fields.It effectively reduces the presence of H_(2)O on the zinc anode surface,minimizing side reactions and inducing oriented growth of Zn(101)crystal plane.Furthermore,a high concentration of MES anions at the interface can effectively prevent the space charge effect caused by the depletion of SO_(4)^(2-)anions,thereby inhibiting dendrite growth caused by the local electric field.This strategy enables Zn//Zn symmetric cells to achieve 3000 h of cycle life and demonstrates excellent performance in high mass-loading,low N/P ratio Zn//VO_(2)full cells.展开更多
Solid-state polymer electrolytes hold the potential for market application due to the combination of advantageous properties,such as flexibility,ease of processing and low cost.However,the sluggish ion transport and p...Solid-state polymer electrolytes hold the potential for market application due to the combination of advantageous properties,such as flexibility,ease of processing and low cost.However,the sluggish ion transport and poor high-voltage stability pose significant challenges for the practical application of polymer-based solid-state lithium metal batteries(SSLMBs).Therefore,the design and development of polymer-based SSLMBs toward fast-charge and high-voltage is of great significance in high-energydensity devices.Herein,this review deeply analyzes the mechanism of ion transport and anti-oxidation of polymer-based solidstate electrolytes.Furthermore,we also systematically and comprehensively summarize the factors that affect ionic conductivity and the electrochemical window.Moreover,we outline the solution strategies for simultaneously enhancing both ionic conductivity and high-voltage stability.Besides,we discuss the main challenges and the future prospects of polymer-based SSLMBs for further studies.It is hoped that this review can provide both advances and fundamentals to the research community and pave the way for the development of SSLMBs.展开更多
Drug discovery is costly and time consuming,and modern drug discovery endeavors are progressively reliant on computational methodologies,aiming to mitigate temporal and financial expenditures associated with the proce...Drug discovery is costly and time consuming,and modern drug discovery endeavors are progressively reliant on computational methodologies,aiming to mitigate temporal and financial expenditures associated with the process.In particular,the time required for vaccine and drug discovery is prolonged during emergency situations such as the coronavirus 2019 pandemic.Recently,the performance of deep learning methods in drug virtual screening has been particularly prominent.It has become a concern for researchers how to summarize the existing deep learning in drug virtual screening,select different models for different drug screening problems,exploit the advantages of deep learning models,and further improve the capability of deep learning in drug virtual screening.This review first introduces the basic concepts of drug virtual screening,common datasets,and data representation methods.Then,large numbers of common deep learning methods for drug virtual screening are compared and analyzed.In addition,a dataset of different sizes is constructed independently to evaluate the performance of each deep learning model for the difficult problem of large-scale ligand virtual screening.Finally,the existing challenges and future directions in the field of virtual screening are presented.展开更多
The construction of robust coupling catalysts for accelerating electrocatalytic oxygen reduction reaction(ORR)through the modulation of the electronic structure and local atomic configuration is critical but remains c...The construction of robust coupling catalysts for accelerating electrocatalytic oxygen reduction reaction(ORR)through the modulation of the electronic structure and local atomic configuration is critical but remains challenging.Herein,we report a facile and effective isolation-polymerization-pyrolysis(IPP)strategy for high-precision synthesis of single-atomic Mn sites coupled with Fe_(3)C nanoparticles encapsulated in N-doped porous carbon matrixes(Mn SAs/Fe_(3)C NPs@NPC)catalyst derived from predesigned bimetallic Fe/Mn polyphthalocyanine(FeMn-BPPc)conjugated polymer networks by solid-phase reaction approach.Benefiting from the synergistic effects between the single-atomic Mn-N_(4)sites and Fe_(3)C NPs as well as the confinement effect of NPC,the Mn SAs/Fe_(3)C NPs@NPC catalyst exhibited excellent electrocatalytic activity and stability for ORR.The assembled Znair battery displayed larger power density of 186 mW·cm^(−2)than that of Pt/C+Ir/C-based battery.It also exhibits excellent stability without obvious voltage change after 106 cycles with 36 h.Combing in-situ Raman spectra with in-situ attenuated total reflectance surface-enhanced infrared absorption spectroscopy(ATR-SEIRAS)characterization results indicated that the Mn-N_(4)site as an active site for the O_(2)adsorption-activation process,which effectively facilitates the generation of key*OOH intermediates and*OH desorption to promote the multielectron reaction kinetics.Theoretical calculation reveals that the excellent electrocatalytic performance originates from the charge redistribution and the d orbital shift resulting from Mn-Fe bond,which buffers the activity of ORR through the electron reservoir capable of electron donation or releasing.This work paves a novel IPP strategy for constructing high-performance coupling electrocatalyst towards the ORR for energy conversion devices.展开更多
Metal-organic frameworks(MOFs)have been used to encapsulate active metal nanoparticles(MNPs)to fabricate MNPs@MOFs composites with high catalytic efficiencies.However,the diffusion of reactants and the accessibility o...Metal-organic frameworks(MOFs)have been used to encapsulate active metal nanoparticles(MNPs)to fabricate MNPs@MOFs composites with high catalytic efficiencies.However,the diffusion of reactants and the accessibility of MNPs located in the center of MOFs may be hindered due to the inherent microporous structures of MOFs,which would affect the catalytic activities of MNPs.Herein,we report a solvent assisted ligand exchange-hydrogen reduction(SALE-HR)strategy to selectively encapsulate ultrafine MNPs(Pd or Pt)within the shallow layers of a MOF,i.e.,UiO-67.The particle sizes of the encapsulated MNPs and the thickness of the MNPs-embedded layers can be adjusted easily by controlling the SALE conditions(e.g.time and temperature).Crucially,the LE-Pd@UiO-80-0.5 composite with the thinnest Pd-embedded layers displays remarkable catalytic efficiency with a high turnover frequency(TOF)value of 600 h^-1towards hydrogenation of nitrobenzene under 1 atm H_2at room temperature.The results indicate that the catalytic efficiency and the utilization of MNPs can be enhanced by compactly encapsulating MNPs within the shallow layers of MOFs as close to their outer surfaces as possible,owing to the short masstransfer distance and enhanced accessibility of overall MNPs.展开更多
Entity resolution （ER） is the problem of identi- fying and grouping different manifestations of the same real world object. Algorithmic approaches have been developed where most tasks offer superior performance unde...Entity resolution （ER） is the problem of identi- fying and grouping different manifestations of the same real world object. Algorithmic approaches have been developed where most tasks offer superior performance under super- vised learning. However, the prohibitive cost of labeling training data is still a huge obstacle for detecting duplicate query records from online sources. Furthermore, the unique combinations of noisy data with missing elements make ER tasks more challenging. To address this, transfer learning has been adopted to adaptively share learned common structures of similarity scoring problems between multiple sources. Al- though such techniques reduce the labeling cost so that it is linear with respect to the number of sources, its random sam- piing strategy is not successful enough to handle the ordinary sample imbalance problem. In this paper, we present a novel multi-source active transfer learning framework to jointly select fewer data instances from all sources to train classi- fiers with constant precision/recall. The intuition behind our approach is to actively label the most informative samples while adaptively transferring collective knowledge between sources. In this way, the classifiers that are learned can be both label-economical and flexible even for imbalanced or quality diverse sources. We compare our method with the state-of-the-art approaches on real-word datasets. Our exper- imental results demonstrate that our active transfer learning algorithm can achieve impressive performance with far fewerlabeled samples for record matching with numerous and var- ied sources.展开更多
Lithium-sulfur batteries(LSBs)are regarded as the most promising next-generation energy system due to their high theoretical energy density.However,LSBs suffer the“shuttle effect”if undergoing the solid-liquid-solid...Lithium-sulfur batteries(LSBs)are regarded as the most promising next-generation energy system due to their high theoretical energy density.However,LSBs suffer the“shuttle effect”if undergoing the solid-liquid-solid sulfur conversion process during cycling.Herein,we design a solvent-in-salt(SIS)electrolyte with co-solvent vinylene carbonate(VC)to synthesize an in situ dense cathode electrolyte interface(CEI)and successfully change sulfur conversion into a solid-solid way to avoid shuttle effect by separating the contact of sulfur and ether solvent.Dense CEI is formed at the beginning of first discharge by the combined action of SIS electrolyte and filmogen VC.Experiments and simulations show that SIS electrolyte controls the initial formed lithium polysulfides(LiPSs)to stay very closely on the cathode surface,and then converts them into a dense CEI film.As a result,Coulombic efficiency(above 99%)and cycling performance of LSBs are improved.Furthermore,the in situ dense CEI can nearly stop the self-discharge of LSBs,and enable the LSBs to work under a pretty lean electrolyte condition.展开更多
With the rapid development of location-based services, a particularly important aspect of start-up marketing research is to explore and characterize points of interest （PoIs） such as restaurants and hotels on maps. ...With the rapid development of location-based services, a particularly important aspect of start-up marketing research is to explore and characterize points of interest （PoIs） such as restaurants and hotels on maps. However, due to the lack of direct access to PoI databases, it is necessary to rely on existing APIs to query Pols within a region and calculate PoI statistics. Unfortunately, public APIs generally im- pose a limit on the maximum number of queries. Therefore, we propose effective and efficient sampling methods based on road networks to sample PoIs on maps and provide unbiased estimators for calculating PoI statistics. In general, the more intense the roads, the denser the distribution of PoIs is within a region. Experimental results show that compared with state-of-the-art methods, our sampling methods improve the efficiency of aggregate statistical estimations.展开更多
Designing a functional occlusal surface for denture crowns is a complex and important task in prosthodontics.Manual design is time-consuming and heavily relies on the dentist's experience,as it requires careful co...Designing a functional occlusal surface for denture crowns is a complex and important task in prosthodontics.Manual design is time-consuming and heavily relies on the dentist's experience,as it requires careful consideration of occlusal function.Due to the limitations of manual design,the field has turned to data-driven methods for occlusal surface design.However,many of these methods neglect critical geometric details,such as normals and curvature,impacting the quality of the occlusal surface.In this paper,we introduce Diff-OSGN,a novel denture crown occlusal surface generation network based on a denoising diffusion model,which focuses on generating the detailed geometric structure of denture crowns.We model the occlusal surface as a geometry map based on the occlusal plane,incorporating height and normal maps rasterized from intra-oral crown scanning.Both maps represent occlusal surface geometry,and their combination further enhances these details.Considering the crucial occlusal information,we extract features from the geometry maps of adjacent and occlusal teeth,using them as conditions in the reverse diffusion process to train our network for optimal occlusal function.Additionally,we define three geometric operators and corresponding loss functions as constraints to better extract geometric features of the target occlusal surface,such as ridges and grooves,for adequate supervision.Our results demonstrate that Diff-OSGN provides quantitatively and qualitatively superior performance than competing baselines and state-of-the-art methods.展开更多
基金supported by the Jiangsu Province Traditional Chinese Medicine Technology Development Plan Project,Nos.MS2023113(to JC),MS2022090Young and Middle-aged Academic Leaders of Jiangsu Qing-Lan Project(to GL).
文摘Unlike mammals,zebrafish possess a remarkable ability to regenerate their spinal cord after injury,making them an ideal vertebrate model for studying regeneration.While previous research has identified key cell types involved in this process,the underlying molecular and cellular mechanisms remain largely unexplored.In this study,we used single-cell RNA sequencing to profile distinct cell populations at different stages of spinal cord injury in zebrafish.Our analysis revealed that multiple subpopulations of neurons showed persistent activation of genes associated with axonal regeneration post injury,while molecular signals promoting growth cone collapse were inhibited.Radial glial cells exhibited significant proliferation and differentiation potential post injury,indicating their intrinsic roles in promoting neurogenesis and axonal regeneration,respectively.Additionally,we found that inflammatory factors rapidly decreased in the early stages following spinal cord injury,creating a microenvironment permissive for tissue repair and regeneration.Furthermore,oligodendrocytes lost maturity markers while exhibiting increased proliferation following injury.These findings demonstrated that the rapid and orderly regulation of inflammation,as well as the efficient proliferation and redifferentiation of new neurons and glial cells,enabled zebrafish to reconstruct the spinal cord.This research provides new insights into the cellular transitions and molecular programs that drive spinal cord regeneration,offering promising avenues for future research and therapeutic strategies.
文摘The development of advanced support is conducive to promoting the practical application of fuel cells but remains an enormous challenge in terms of stabilizing catalyst particles and enabling improved accessibility to O_(2).Beyond solid carbon black and conventional porous carbon,we demonstrate a new type of mesoporous bowl-like carbon support with a high specific surface area of over 1200 m^(2)g^(-1)and~4 nm pore.Both rotating disk electrode and membrane electrode assembly(MEA)tests show that BC-supported Pt_(3)Co(Pt_(3)Co/BC)catalyst greatly outperforms hollow porous carbon spheres and solid carbon spheres supported Pt_(3)Co catalysts.The Pt_(3)Co/BC catalysts exhibited remarkable performance as cathode catalyst in MEA,achieving a comparable open-circuit voltage of approximately 0.95 V under H_(2)-air condition and a current density of 1.3 A cm^(-2)at 0.6 V with a loading of only 0.2 mgPt cm^(-2).Diffusion simulations and physical characterizations demonstrate that the high porosity and highly accessible pore structure of BC support facilitate the uniform distribution of catalyst particles and enhance the mass transport of O_(2),thereby resulting in a significant improvement in catalytic activity and durability.This work provides new insights into the influence of support shape on mass transport of reactants and electrochemical performances of the catalyst in MEA.
文摘The development of efficient and cost‐effective metal‐free electrocatalysts for oxygen reduction reaction(ORR)has become crucial for electrochemical energy systems.However,reasonably validating and precisely regulating the active sites for designing optimized materials are still challenging.Herein,we report a precise and controllable tandem strategy to boost the ORR activity based on metal‐free covalent organic frameworks(MFCOFs)comprising imine‐N,thiophene‐S,or triazine‐N.Among these MFCOFs,post‐tandem BTT‐TAT‐COF structure displayed a more positive catalytic capability and excellent electrochemical stability,indicating that the synergistic catalysis of multiple active sites induced the ORR catalytic activity through the conjugated skeleton of the structure.Density‐functional theory calculations suggest that the series‐connected backbone contained highly effective electrocatalytic active centers and provided synergistic catalysis.More importantly,this strategy highlights new opportunities for the advancement of efficient COF‐based metal‐free ORR catalysts.
文摘The electronic conductivity and durability of porous carbon supports can be improved by increasing the degree of graphitization in the material;however,the preparation of highly graphitic porous carbon using conventional furnaces remains a significant challenge.Herein,we demonstrate a universal and highly efficient carbothermal shock strategy that significantly improves the degree of graphitization of porous carbon supports,including bowl‐like carbon,hollow carbon spheres,ZIF8‐derived carbon,BP2000,and Ketjen EC 300J.Taking bowl‐like carbon as an example,we illustrate the synthesis of a graphitized bowl‐like carbon(G‐BC‐S)support and evaluate the performance of PtCo/G‐BC‐S in the oxygen reduction reaction(ORR)in rotating disk electrodes(RDE)and H_(2)/air PEM single cells.PtCo/G‐BC‐S exhibits faster ORR kinetics than PtCo/BC and Pt/C,with little loss of activity(25%)and only 13 mV of E1/2 decay after 20000 cycles accelerated stress testing under 1.0-1.5 V vs.a reversible hydrogen electrode(RHE).The significantly enhanced performance of the PtCo/G‐BC‐S catalyst arises from the high activity and chemical/structural stability of the PtCo intermetallic nanoparticles and from the high degree of graphitization and well‐defined porous structure of the bowl‐like carbon support,which confers excellent electrical conductivity and oxygen transport properties.This study provides a reliable and universal strategy for the development of high‐performance porous carbon supports for practical applications in fuel cells.
基金supported by the National Science Foundation of China(NSFC)underGrants 61876217 and 62176175the Innovative Team of Jiangsu Province under Grant XYDXX-086Jiangsu Postgraduate Research and Innovation Plan(KYCX20_2762).
文摘Interpreting deep neural networks is of great importance to understand and verify deep models for natural language processing(NLP)tasks.However,most existing approaches only focus on improving the performance of models but ignore their interpretability.In this work,we propose a Randomly Wired Graph Neural Network(RWGNN)by using graph to model the structure of Neural Network,which could solve two major problems(word-boundary ambiguity and polysemy)of ChineseNER.Besides,we develop a pipeline to explain the RWGNNby using Saliency Map and Adversarial Attacks.Experimental results demonstrate that our approach can identify meaningful and reasonable interpretations for hidden states of RWGNN.
基金supported by the National Natural Science Foundation of China(Grant Nos.81771319,82202436)the Medical Research Project of Jiangsu Commission of Health(Grant No.ZDB2020004)+1 种基金the Scientific Research Project of Nantong Municipal Health Commission(Grant No.MA2021016)The First People’s Hospital of Nantong Provincial and Ministerial High-Level Science and Technology Project Cultivation Fund(Grant No.YPYJJZD009).
文摘Spinal cord injury(SCI),a complex neurological disorder,triggers a series of devastating neuropathological events such as ischemia,oxidative stress,inflammatory events,neuronal apoptosis,and motor dysfunction.However,the classical necrosome,which consists of receptor-interacting protein(RIP)1,RIP3,and mixed-lineage kinase domain-like protein,is believed to control a novel type of programmed cell death called necroptosis,through tumour necrosis factor-alpha/tumour necrosis factor receptor-1 signalling or other stimuli.Several studies reported that necroptosis plays an important role in neural cell damage,release of intracellular pro-inflammatory factors,lysosomal dysfunction and endoplasmic reticulum stress.Recent research indicates that necroptosis is crucial to the pathophysiology of a number of neurological disorders and SCIs.In our review,we summarize the potential role of programmed cell death regulated by necroptosis in SCI based on its molecular and pathophysiological mechanisms.We also summarize the targets of several necroptosis pathways,which provide a more reliable reference for the treatment of SCI.
基金This research has been supported by the National Natural Science Foundation of China under grant 61728205,61673290,61672371,61750110534,61876217Science&Technology Development Project of Suzhou under grant SYG201817.
文摘The pattern of thematic progression,reflecting the semantic relationships between contextual two sentences,is an important subject in discourse analysis.We introduce a new corpus of Chinese news discourses annotated with thematic progression information and explore some computational methods to automatically extracting the discourse structural features of simplified thematic progression pattern(STPP)between contextual sentences in a text.Furthermore,these features are used in a hybrid approach to a major discourse analysis task,Chinese coreference resolution.This novel approach is built up via heuristic sieves and a machine learning method that comprehensively utilizes both the top-down STPP features and the bottom-up semantic features.Experimental results on the intersection of the CoNLL-2012 task shared dataset and the CDTC corpus demonstrate the effectiveness of our proposed approach.
基金supported by the National Natural Science Foundation of China(Nos.22325802 and U22A20417)the Natural Science Foundation of Guangdong Province(No.2024A1515012535)+2 种基金the Science and Technology Program of Guangzhou(No.2023B03J1281)the Guangdong Basic and Applied Basic Research Foundation(No.2023B1515120005)the State Key Laboratory of Catalysis(No.2024SKL-A-009).
文摘The differences in Zn crystal plane kinetics can lead to non-uniform deposition,promoting dendrite growth and side reactions,especially under high deposition capacities.Fast kinetics can also cause anion depletion on the zinc anode surface,leading to uneven electric field distribution and worsening these issues.Inducing preferred electrodeposition of the Zn(101)crystal plane can ensure dense epitaxial growth and achieve fast reaction kinetics.However,its highly reactive and wave-like arrangement will also lead to higher hydrogen evolution activity and cause uneven electric field distribution,accelerating side reactions and dendrite growth.This study utilizes the adsorption ability of 2-mercaptoethanesulfonate(MES)anion on the zinc anode surface to optimize the interfacial concentration and electric fields.It effectively reduces the presence of H_(2)O on the zinc anode surface,minimizing side reactions and inducing oriented growth of Zn(101)crystal plane.Furthermore,a high concentration of MES anions at the interface can effectively prevent the space charge effect caused by the depletion of SO_(4)^(2-)anions,thereby inhibiting dendrite growth caused by the local electric field.This strategy enables Zn//Zn symmetric cells to achieve 3000 h of cycle life and demonstrates excellent performance in high mass-loading,low N/P ratio Zn//VO_(2)full cells.
基金supported by the National Natural Science Foundation of China(22272058,22072048)the Key Technologies R&D Program of Guangdong Province(2023B0909060003)the Guangdong Basic and Applied Basic Research Foundation(2022A1515111097)。
文摘Solid-state polymer electrolytes hold the potential for market application due to the combination of advantageous properties,such as flexibility,ease of processing and low cost.However,the sluggish ion transport and poor high-voltage stability pose significant challenges for the practical application of polymer-based solid-state lithium metal batteries(SSLMBs).Therefore,the design and development of polymer-based SSLMBs toward fast-charge and high-voltage is of great significance in high-energydensity devices.Herein,this review deeply analyzes the mechanism of ion transport and anti-oxidation of polymer-based solidstate electrolytes.Furthermore,we also systematically and comprehensively summarize the factors that affect ionic conductivity and the electrochemical window.Moreover,we outline the solution strategies for simultaneously enhancing both ionic conductivity and high-voltage stability.Besides,we discuss the main challenges and the future prospects of polymer-based SSLMBs for further studies.It is hoped that this review can provide both advances and fundamentals to the research community and pave the way for the development of SSLMBs.
基金the National Natural Science Foundation of China(62073231,62176175,62172076)National Research Project(2020YFC2006602)+2 种基金Provincial Key Laboratory for Computer Information Processing Technology,Soochow University(KJS2166)Opening Topic Fund of Big Data Intelligent Engineering Laboratory of Jiangsu Province(SDGC2157)Postgraduate Research&Practice Innovation Program of Jiangsu Province.
文摘Drug discovery is costly and time consuming,and modern drug discovery endeavors are progressively reliant on computational methodologies,aiming to mitigate temporal and financial expenditures associated with the process.In particular,the time required for vaccine and drug discovery is prolonged during emergency situations such as the coronavirus 2019 pandemic.Recently,the performance of deep learning methods in drug virtual screening has been particularly prominent.It has become a concern for researchers how to summarize the existing deep learning in drug virtual screening,select different models for different drug screening problems,exploit the advantages of deep learning models,and further improve the capability of deep learning in drug virtual screening.This review first introduces the basic concepts of drug virtual screening,common datasets,and data representation methods.Then,large numbers of common deep learning methods for drug virtual screening are compared and analyzed.In addition,a dataset of different sizes is constructed independently to evaluate the performance of each deep learning model for the difficult problem of large-scale ligand virtual screening.Finally,the existing challenges and future directions in the field of virtual screening are presented.
基金State Key Laboratory of Catalytic Materials and Reaction Engineering(RIPP,SINOPEC)Taishan Scholars Program of Shandong Province(No.tsqn201909065)+5 种基金Shandong Provincial Natural Science Foundation(Nos.ZR2021YQ15,ZR2020QB174,and ZR2019MB022)the National Natural Science Foundation of China(Nos.22108306 and 21902182)the Fundamental Research Funds for the Central Universities(Nos.2022YQHH01 and 22CX07009A)the State Key Laboratory of Organic-Inorganic Composites(No.oic202101006)Post-graduate Innovation Fund of China University of Petroleum(East China)(No.YCX2021064)the Research Fund Program of Key Laboratory of Fuel Cell Technology of Guangdong Province,the Key Laboratory of Advanced Energy Materials Chemistry(Ministry of Education),and the Key Laboratory of Functional Inorganic Material Chemistry(Heilongjiang University),Ministry of Education.
文摘The construction of robust coupling catalysts for accelerating electrocatalytic oxygen reduction reaction(ORR)through the modulation of the electronic structure and local atomic configuration is critical but remains challenging.Herein,we report a facile and effective isolation-polymerization-pyrolysis(IPP)strategy for high-precision synthesis of single-atomic Mn sites coupled with Fe_(3)C nanoparticles encapsulated in N-doped porous carbon matrixes(Mn SAs/Fe_(3)C NPs@NPC)catalyst derived from predesigned bimetallic Fe/Mn polyphthalocyanine(FeMn-BPPc)conjugated polymer networks by solid-phase reaction approach.Benefiting from the synergistic effects between the single-atomic Mn-N_(4)sites and Fe_(3)C NPs as well as the confinement effect of NPC,the Mn SAs/Fe_(3)C NPs@NPC catalyst exhibited excellent electrocatalytic activity and stability for ORR.The assembled Znair battery displayed larger power density of 186 mW·cm^(−2)than that of Pt/C+Ir/C-based battery.It also exhibits excellent stability without obvious voltage change after 106 cycles with 36 h.Combing in-situ Raman spectra with in-situ attenuated total reflectance surface-enhanced infrared absorption spectroscopy(ATR-SEIRAS)characterization results indicated that the Mn-N_(4)site as an active site for the O_(2)adsorption-activation process,which effectively facilitates the generation of key*OOH intermediates and*OH desorption to promote the multielectron reaction kinetics.Theoretical calculation reveals that the excellent electrocatalytic performance originates from the charge redistribution and the d orbital shift resulting from Mn-Fe bond,which buffers the activity of ORR through the electron reservoir capable of electron donation or releasing.This work paves a novel IPP strategy for constructing high-performance coupling electrocatalyst towards the ORR for energy conversion devices.
基金the National Natural Science Foundation of China(21825802,21908068)the Fundamental Research Funds for the Central Universities(2019PY11,2019MS041)+2 种基金the Science and Technology Program of Guangzhou(201804020009)the State Key Laboratory of Pulp and Paper Engineering(2017ZD04,2018TS03)the Natural Science Foundation of Guangdong Province(2016A050502004,2017A030312005,2020A1515010376)。
文摘Metal-organic frameworks(MOFs)have been used to encapsulate active metal nanoparticles(MNPs)to fabricate MNPs@MOFs composites with high catalytic efficiencies.However,the diffusion of reactants and the accessibility of MNPs located in the center of MOFs may be hindered due to the inherent microporous structures of MOFs,which would affect the catalytic activities of MNPs.Herein,we report a solvent assisted ligand exchange-hydrogen reduction(SALE-HR)strategy to selectively encapsulate ultrafine MNPs(Pd or Pt)within the shallow layers of a MOF,i.e.,UiO-67.The particle sizes of the encapsulated MNPs and the thickness of the MNPs-embedded layers can be adjusted easily by controlling the SALE conditions(e.g.time and temperature).Crucially,the LE-Pd@UiO-80-0.5 composite with the thinnest Pd-embedded layers displays remarkable catalytic efficiency with a high turnover frequency(TOF)value of 600 h^-1towards hydrogenation of nitrobenzene under 1 atm H_2at room temperature.The results indicate that the catalytic efficiency and the utilization of MNPs can be enhanced by compactly encapsulating MNPs within the shallow layers of MOFs as close to their outer surfaces as possible,owing to the short masstransfer distance and enhanced accessibility of overall MNPs.
文摘Entity resolution （ER） is the problem of identi- fying and grouping different manifestations of the same real world object. Algorithmic approaches have been developed where most tasks offer superior performance under super- vised learning. However, the prohibitive cost of labeling training data is still a huge obstacle for detecting duplicate query records from online sources. Furthermore, the unique combinations of noisy data with missing elements make ER tasks more challenging. To address this, transfer learning has been adopted to adaptively share learned common structures of similarity scoring problems between multiple sources. Al- though such techniques reduce the labeling cost so that it is linear with respect to the number of sources, its random sam- piing strategy is not successful enough to handle the ordinary sample imbalance problem. In this paper, we present a novel multi-source active transfer learning framework to jointly select fewer data instances from all sources to train classi- fiers with constant precision/recall. The intuition behind our approach is to actively label the most informative samples while adaptively transferring collective knowledge between sources. In this way, the classifiers that are learned can be both label-economical and flexible even for imbalanced or quality diverse sources. We compare our method with the state-of-the-art approaches on real-word datasets. Our exper- imental results demonstrate that our active transfer learning algorithm can achieve impressive performance with far fewerlabeled samples for record matching with numerous and var- ied sources.
基金supported by the National Science Foundation of China(No.21776105)the Natural Science Foundation of Guangdong Province(No.2019A1515011720)Science and Technology Program of Guangzhou(No.201904010340).
文摘Lithium-sulfur batteries(LSBs)are regarded as the most promising next-generation energy system due to their high theoretical energy density.However,LSBs suffer the“shuttle effect”if undergoing the solid-liquid-solid sulfur conversion process during cycling.Herein,we design a solvent-in-salt(SIS)electrolyte with co-solvent vinylene carbonate(VC)to synthesize an in situ dense cathode electrolyte interface(CEI)and successfully change sulfur conversion into a solid-solid way to avoid shuttle effect by separating the contact of sulfur and ether solvent.Dense CEI is formed at the beginning of first discharge by the combined action of SIS electrolyte and filmogen VC.Experiments and simulations show that SIS electrolyte controls the initial formed lithium polysulfides(LiPSs)to stay very closely on the cathode surface,and then converts them into a dense CEI film.As a result,Coulombic efficiency(above 99%)and cycling performance of LSBs are improved.Furthermore,the in situ dense CEI can nearly stop the self-discharge of LSBs,and enable the LSBs to work under a pretty lean electrolyte condition.
基金This work was partially supported by the National Natural Science Foundation of China （NSFC） （Grant N os. 61170020, 61402311, 61440053）, and the US National Science Foundation （IIS- 1115417）.
文摘With the rapid development of location-based services, a particularly important aspect of start-up marketing research is to explore and characterize points of interest （PoIs） such as restaurants and hotels on maps. However, due to the lack of direct access to PoI databases, it is necessary to rely on existing APIs to query Pols within a region and calculate PoI statistics. Unfortunately, public APIs generally im- pose a limit on the maximum number of queries. Therefore, we propose effective and efficient sampling methods based on road networks to sample PoIs on maps and provide unbiased estimators for calculating PoI statistics. In general, the more intense the roads, the denser the distribution of PoIs is within a region. Experimental results show that compared with state-of-the-art methods, our sampling methods improve the efficiency of aggregate statistical estimations.
基金supported by the National Natural Science Foundation of China(62172257)the Natural Science Foundation of Shandong Province(Major Basic Research)project(ZR2024ZD12)the HPC Platform of ShanghaiTech University.
文摘Designing a functional occlusal surface for denture crowns is a complex and important task in prosthodontics.Manual design is time-consuming and heavily relies on the dentist's experience,as it requires careful consideration of occlusal function.Due to the limitations of manual design,the field has turned to data-driven methods for occlusal surface design.However,many of these methods neglect critical geometric details,such as normals and curvature,impacting the quality of the occlusal surface.In this paper,we introduce Diff-OSGN,a novel denture crown occlusal surface generation network based on a denoising diffusion model,which focuses on generating the detailed geometric structure of denture crowns.We model the occlusal surface as a geometry map based on the occlusal plane,incorporating height and normal maps rasterized from intra-oral crown scanning.Both maps represent occlusal surface geometry,and their combination further enhances these details.Considering the crucial occlusal information,we extract features from the geometry maps of adjacent and occlusal teeth,using them as conditions in the reverse diffusion process to train our network for optimal occlusal function.Additionally,we define three geometric operators and corresponding loss functions as constraints to better extract geometric features of the target occlusal surface,such as ridges and grooves,for adequate supervision.Our results demonstrate that Diff-OSGN provides quantitatively and qualitatively superior performance than competing baselines and state-of-the-art methods.
