The growing demand for wireless connectivity has made massive multiple-input multiple-output(MIMO)a cornerstone of modern communication systems.To optimize network performance and resource allocation,an efficient and ...The growing demand for wireless connectivity has made massive multiple-input multiple-output(MIMO)a cornerstone of modern communication systems.To optimize network performance and resource allocation,an efficient and robust approach is joint device activity detection and channel estimation.In this paper,we present an approach utilizing score-based generative models to address the underdetermined nature of channel estimation,which is data-driven and well-suited for the complex and dynamic environment of massive MIMO systems.Our experimental results,based on a comprehensive dataset generated through Monte-Carlo sampling,demonstrate the high precision of our channel estimation approach,with errors reduced to as low as-45 d B,and exceptional accuracy in detecting active devices.展开更多
BACKGROUND Patients with chronic obstructive pulmonary disease(COPD)frequently experience exacerbations requiring multiple hospitalizations over prolonged disease courses,which predispose them to generalized anxiety d...BACKGROUND Patients with chronic obstructive pulmonary disease(COPD)frequently experience exacerbations requiring multiple hospitalizations over prolonged disease courses,which predispose them to generalized anxiety disorder(GAD).This comorbidity exacerbates breathing difficulties,activity limitations,and social isolation.While previous studies predominantly employed the GAD 7-item scale for screening,this approach is somewhat subjective.The current literature on predictive models for GAD risk in patients with COPD is limited.AIM To construct and validate a GAD risk prediction model to aid healthcare professionals in preventing the onset of GAD.METHODS This retrospective analysis encompassed patients with COPD treated at our institution from July 2021 to February 2024.The patients were categorized into a modeling(MO)group and a validation(VA)group in a 7:3 ratio on the basis of the occurrence of GAD.Univariate and multivariate logistic regression analyses were utilized to construct the risk prediction model,which was visualized using forest plots.The model’s performance was evaluated using Hosmer-Lemeshow(H-L)goodness-of-fit test and receiver operating characteristic(ROC)curve analysis.RESULTS A total of 271 subjects were included,with 190 in the MO group and 81 in the VA group.GAD was identified in 67 patients with COPD,resulting in a prevalence rate of 24.72%(67/271),with 49 cases(18.08%)in the MO group and 18 cases(22.22%)in the VA group.Significant differences were observed between patients with and without GAD in terms of educational level,average household income,smoking history,smoking index,number of exacerbations in the past year,cardiovascular comorbidities,disease knowledge,and personality traits(P<0.05).Multivariate logistic regression analysis revealed that lower education levels,household income<3000 China yuan,smoking history,smoking index≥400 cigarettes/year,≥two exacerbations in the past year,cardiovascular comorbidities,complete lack of disease information,and introverted personality were significant risk factors for GAD in the MO group(P<0.05).ROC analysis indicated that the area under the curve for predicting GAD in the MO and VA groups was 0.978 and 0.960.The H-L test yieldedχ^(2) values of 6.511 and 5.179,with P=0.275 and 0.274.Calibration curves demonstrated good agreement between predicted and actual GAD occurrence risks.CONCLUSION The developed predictive model includes eight independent risk factors:Educational level,household income,smoking history,smoking index,number of exacerbations in the past year,presence of cardiovascular comorbidities,level of disease knowledge,and personality traits.This model effectively predicts the onset of GAD in patients with COPD,enabling early identification of high-risk individuals and providing a basis for early preventive interventions by nursing staff.展开更多
Panoramic images, offering a 360-degree view, are essential in virtual reality(VR) and augmented reality(AR), enhancing realism with high-quality textures. However, acquiring complete and high-quality panoramic textur...Panoramic images, offering a 360-degree view, are essential in virtual reality(VR) and augmented reality(AR), enhancing realism with high-quality textures. However, acquiring complete and high-quality panoramic textures is challenging. This paper introduces a method using generative adversarial networks(GANs) and the contrastive language-image pretraining(CLIP) model to restore and control texture in panoramic images. The GAN model captures complex structures and maintains consistency, while CLIP enables fine-grained texture control via semantic text-image associations. GAN inversion optimizes latent codes for precise texture details. The resulting low dynamic range(LDR) images are converted to high dynamic range(HDR) using the Blender engine for seamless texture blending. Experimental results demonstrate the effectiveness and flexibility of this method in panoramic texture restoration and generation.展开更多
Large language models(LLMs)have undergone significant expansion and have been increasingly integrated across various domains.Notably,in the realm of robot task planning,LLMs harness their advanced reasoning and langua...Large language models(LLMs)have undergone significant expansion and have been increasingly integrated across various domains.Notably,in the realm of robot task planning,LLMs harness their advanced reasoning and language comprehension capabilities to formulate precise and efficient action plans based on natural language instructions.However,for embodied tasks,where robots interact with complex environments,textonly LLMs often face challenges due to a lack of compatibility with robotic visual perception.This study provides a comprehensive overview of the emerging integration of LLMs and multimodal LLMs into various robotic tasks.Additionally,we propose a framework that utilizes multimodal GPT-4V to enhance embodied task planning through the combination of natural language instructions and robot visual perceptions.Our results,based on diverse datasets,indicate that GPT-4V effectively enhances robot performance in embodied tasks.This extensive survey and evaluation of LLMs and multimodal LLMs across a variety of robotic tasks enriches the understanding of LLM-centric embodied intelligence and provides forward-looking insights towards bridging the gap in Human-Robot-Environment interaction.展开更多
This study explores a novel educational model of generative AI-empowered interdisciplinary project-based learning(PBL).By analyzing the current applications of generative AI technology in information technology curric...This study explores a novel educational model of generative AI-empowered interdisciplinary project-based learning(PBL).By analyzing the current applications of generative AI technology in information technology curricula,it elucidates its advantages and operational mechanisms in interdisciplinary PBL.Combining case studies and empirical research,the investigation proposes implementation pathways and strategies for the generative AI-enhanced interdisciplinary PBL model,detailing specific applications across three phases:project preparation,implementation,and evaluation.The research demonstrates that generative AI-enabled interdisciplinary project-based learning can effectively enhance students’learning motivation,interdisciplinary thinking capabilities,and innovative competencies,providing new conceptual frameworks and practical approaches for educational model innovation.展开更多
This critical review provides an in-depth analysis of Large Language Models(LLMs),encompassing their foundational principles,diverse applications,and advanced training methodologies.We critically examine the evolution...This critical review provides an in-depth analysis of Large Language Models(LLMs),encompassing their foundational principles,diverse applications,and advanced training methodologies.We critically examine the evolution from Recurrent Neural Networks(RNNs)to Transformer models,highlighting the significant advancements and innovations in LLM architectures.The review explores state-of-the-art techniques such as in-context learning and various fine-tuning approaches,with an emphasis on optimizing parameter efficiency.We also discuss methods for aligning LLMs with human preferences,including reinforcement learning frameworks and human feedback mechanisms.The emerging technique of retrieval-augmented generation,which integrates external knowledge into LLMs,is also evaluated.Additionally,we address the ethical considerations of deploying LLMs,stressing the importance of responsible and mindful application.By identifying current gaps and suggesting future research directions,this review provides a comprehensive and critical overview of the present state and potential advancements in LLMs.This work serves as an insightful guide for researchers and practitioners in artificial intelligence,offering a unified perspective on the strengths,limitations,and future prospects of LLMs.展开更多
Objective:Generative artificial intelligence(AI)technology,represented by large language models(LLMs),has gradually been developed for traditional Chinese medicine(TCM);however,challenges remain in effectively enhanci...Objective:Generative artificial intelligence(AI)technology,represented by large language models(LLMs),has gradually been developed for traditional Chinese medicine(TCM);however,challenges remain in effectively enhancing AI applications for TCM.Therefore,this study is the first systematic review to analyze LLMs in TCM retrospectively,focusing on and summarizing the evidence of their performance in generative tasks.Methods:We extensively searched electronic databases for articles published until June 2024 to identify publicly available studies on LLMs in TCM.Two investigators independently selected and extracted the related information and evaluation metrics.Based on the available data,this study used descriptive analysis for a comprehensive systematic review of LLM technology related to TCM.Results:Ten studies published between 2023 and 2024 met our eligibility criteria and were included in this review,including 40%LLMs in the TCM vertical domain,40%containing TCM data,and 20%honoring the TCM contribution,with a foundational model parameter range from 1.8 to 33 billion.All included studies used manual or automatic evaluation metrics to evaluate model performance and fully discussed the challenges and contributions through an overview of LLMs in TCM.Conclusions:LLMs have achieved significant advantages in TCM applications and can effectively address intelligent TCM tasks.Further in-depth development of LLMs is needed in various vertical TCM fields,including clinical and fundamental research.Focusing on the functional segmentation development direction of generative AI technologies in TCM application scenarios to meet the practical needs-oriented demands of TCM digitalization is essential.展开更多
In order to detect whether the data conforms to the given model, it is necessary to diagnose the data in the statistical way. The diagnostic problem in generalized nonlinear models based on the maximum Lq-likelihood e...In order to detect whether the data conforms to the given model, it is necessary to diagnose the data in the statistical way. The diagnostic problem in generalized nonlinear models based on the maximum Lq-likelihood estimation is considered. Three diagnostic statistics are used to detect whether the outliers exist in the data set. Simulation results show that when the sample size is small, the values of diagnostic statistics based on the maximum Lq-likelihood estimation are greater than the values based on the maximum likelihood estimation. As the sample size increases, the difference between the values of the diagnostic statistics based on two estimation methods diminishes gradually. It means that the outliers can be distinguished easier through the maximum Lq-likelihood method than those through the maximum likelihood estimation method.展开更多
Deep learning(DL)has proven to be important for computed tomography(CT)image denoising.However,such models are usually trained under supervision,requiring paired data that may be difficult to obtain in practice.Diffus...Deep learning(DL)has proven to be important for computed tomography(CT)image denoising.However,such models are usually trained under supervision,requiring paired data that may be difficult to obtain in practice.Diffusion models offer unsupervised means of solving a wide range of inverse problems via posterior sampling.In particular,using the estimated unconditional score function of the prior distribution,obtained via unsupervised learning,one can sample from the desired posterior via hijacking and regularization.However,due to the iterative solvers used,the number of function evaluations(NFE)required may be orders of magnitudes larger than for single-step samplers.In this paper,we present a novel image denoising technique for photon-counting CT by extending the unsupervised approach to inverse problem solving to the case of Poisson flow generative models(PFGM)++.By hijacking and regularizing the sampling process we obtain a single-step sampler,that is NFE=1.Our proposed method incorporates posterior sampling using diffusion models as a special case.We demonstrate that the added robustness afforded by the PFGM++framework yields significant performance gains.Our results indicate competitive performance compared to popular supervised,including state-of-the-art diffusion-style models with NFE=1(consistency models),unsupervised,and non-DL-based image denoising techniques,on clinical low-dose CT data and clinical images from a prototype photon-counting CT system developed by GE HealthCare.展开更多
Soil undergoes both elastic and plastic deformations under different loading conditions. A relatively accurate constitutive model of soil behaviors should be capable of predicting the elastic and plastic deformations ...Soil undergoes both elastic and plastic deformations under different loading conditions. A relatively accurate constitutive model of soil behaviors should be capable of predicting the elastic and plastic deformations properly. Among a large number of elastoplastic constitutive models developed over the last several decades, constitutive models based on generalized plasticity have been successfully utilized in modeling the mechanical behavior of various soils. This paper attempts to present a review of the most recent developments of generalized plasticity models for geotechnical problems. After a brief review of generalized plasticity theories and constitutive models, limitations of the original Pastor-Zienkiewicz model in practical application are summarized. Afterwards, recent achievements in the generalized plasticity models for both saturated and unsaturated soils and their applicability are analyzed, and a general approach for modification of generalized plasticity models is highlighted.展开更多
This study aims to provide a predictive vegetation mapping approach based on the spectral data, DEM and Generalized Additive Models (GAMs). GAMs were used as a prediction tool to describe the relationship between vege...This study aims to provide a predictive vegetation mapping approach based on the spectral data, DEM and Generalized Additive Models (GAMs). GAMs were used as a prediction tool to describe the relationship between vegetation and environmental variables, as well as spectral variables. Based on the fitted GAMs model, probability map of species occurrence was generated and then vegetation type of each grid was defined according to the probability of species occurrence. Deviance analysis was employed to test the goodness of curve fitting and drop contribution calculation was used to evaluate the contribution of each predictor in the fitted GAMs models. Area under curve (AUC) of Receiver Operating Characteristic (ROC) curve was employed to assess the results maps of probability. The results showed that: 1) AUC values of the fitted GAMs models are very high which proves that integrating spectral data and environmental variables based on the GAMs is a feasible way to map the vegetation. 2) Prediction accuracy varies with plant community, and community with dense cover is better predicted than sparse plant community. 3) Both spectral variables and environmental variables play an important role in mapping the vegetation. However, the contribution of the same predictor in the GAMs models for different plant communities is different. 4) Insufficient resolution of spectral data, environmental data and confounding effects of land use and other variables which are not closely related to the environmental conditions are the major causes of imprecision.展开更多
The catch and effort data of Sillago sihama fishery in Pakistani waters were used to investigate the performance of two closely related stock assessment models: logistic and generalized surplus-production models. Comp...The catch and effort data of Sillago sihama fishery in Pakistani waters were used to investigate the performance of two closely related stock assessment models: logistic and generalized surplus-production models. Compared with the generalized production model, the logistic model produced more reasonable estimates for parameters such as maximum sustainable yield. The Akaike's Information Criterion values estimated at 4.265 and -51.152 respectively by the logistic and generalized models. Simulation analyses of the S. sihama fishery showed that the estimated and observed abundance indices for the logistic model were closer than those for the generalized production model. Standardized residuals were distributed closer for logistic model, but exhibited a slightly increasing trend for the generalized model. Statistical outliers were seen in 1989 and 1993 for the logistic model, and in 1981 and 1999 for the generalized model. Simulated results revealed that the logistic estimates were close to the true value for low CVs (coefficients of variation) but widely dispersed for high CVs. In contrast, the generalized model estimates were loose for all CV levels. The estimated production model curve parameter was not reasonable at all the tested levels of white noise. With the increase in white noise R2 for the catch per unit effort decreased. Therefore, we conclude that the logistic model performs more reasonably than the generalized production model.展开更多
Objective: To analyze longitudinal binary data by using generalized linear models. The correlation between repeated measures were considered. The general method for analyzing longitudinal binary data was given. Method...Objective: To analyze longitudinal binary data by using generalized linear models. The correlation between repeated measures were considered. The general method for analyzing longitudinal binary data was given. Methods: Generalized estimating equations (GEE) proposed by Zeger and Liang was used. For sevens covariance structures, one method was given for estimating regression and correlation parameters. Results: Regression and coerelation parameters were estimated simultaneously. A Set of program was finished and an example was illustrated. Conclusion: Longitudinal dsta often occur in medical researches and clinical trials. For solving the problem of correlation between repeated measures, it is necessary to use some special methods to cope with this Kind of data.展开更多
There are typical ecosystems of littoral wetlands in the Yellow River Delta.In order to study the relationships between Tamarix chinensis and environmental variables and to predict T.chinensis potential distribution i...There are typical ecosystems of littoral wetlands in the Yellow River Delta.In order to study the relationships between Tamarix chinensis and environmental variables and to predict T.chinensis potential distribution in the Yellow River Delta,641 vegetation samples and 964 soil samples were collected in the area in October of 2004,2005,2006 and 2007.The contents of soil organic matter,total phosphorus,salt,and soluble potassium were determined.Then,the analyzed data were interpolated into spatial raster data by Kriging interpolation method.Meanwhile,the digital elevation model,soil type map and landform unit map of the Yellow River Delta were also collected.Generalized Additive Models(GAMs) were employed to build species-environment model and then simulate the potential distribution of T.chinensis.The results indicated that the distribution of T.chinensis was mainly limited by soil salt content,total soil phosphorus content,soluble potassium content,soil type,landform unit,and elevation.The distribution probability of T.chinensis was produced with a lookup table generated by Grasp Module(based on GAMs) in software ArcView GIS 3.2.The AUC(Area Under Curve) value of validation and cross-validation of ROC(Receive Operating Characteristic) were both higher than 0.8,which suggested that the established model had a high precision for predicting species distribution.展开更多
On foe basis of the Kirchoff-Karman hypothses for the nonlinear bending of thin plates, the three kinds of boundary value problems of nonlinear analysis for perforated fhin plates are presented under the differenr in...On foe basis of the Kirchoff-Karman hypothses for the nonlinear bending of thin plates, the three kinds of boundary value problems of nonlinear analysis for perforated fhin plates are presented under the differenr in-plane boundary conditions and the corresponding generalized varialional principles are established. One can see that all mathematical models presented in this paper are completely new ones and differ from the ordinary von Karman theory. These mathematical models can be applied to the nonlinear analysis and the Stability analysis of perforaled thin plates in arbitraryplane boundary conditions.展开更多
Fault monitoring of bioprocess is important to ensure safety of a reactor and maintain high quality of products. It is difficult to build an accurate mechanistic model for a bioprocess, so fault monitoring based on ri...Fault monitoring of bioprocess is important to ensure safety of a reactor and maintain high quality of products. It is difficult to build an accurate mechanistic model for a bioprocess, so fault monitoring based on rich historical or online database is an effective way. A group of data based on bootstrap method could be resampling stochastically, improving generalization capability of model. In this paper, online fault monitoring of generalized additive models （GAMs） combining with bootstrap is proposed for glutamate fermentation process. GAMs and bootstrap are first used to decide confidence interval based on the online and off-line normal sampled data from glutamate fermentation experiments. Then GAMs are used to online fault monitoring for time, dissolved oxygen, oxygen uptake rate, and carbon dioxide evolution rate. The method can provide accurate fault alarm online and is helpful to provide useful information for removing fault and abnormal phenomena in the fermentation.展开更多
We study the quasi likelihood equation in Generalized Linear Models（GLM） with adaptive design ∑（i=1）^n xi（yi-h（x＇iβ））=0, where yi is a q=vector, and xi is a p×q random matrix. Under some assumptions, i...We study the quasi likelihood equation in Generalized Linear Models（GLM） with adaptive design ∑（i=1）^n xi（yi-h（x＇iβ））=0, where yi is a q=vector, and xi is a p×q random matrix. Under some assumptions, it is shown that the Quasi- Likelihood equation for the GLM has a solution which is asymptotic normal.展开更多
Count data that exhibit over dispersion (variance of counts is larger than its mean) are commonly analyzed using discrete distributions such as negative binomial, Poisson inverse Gaussian and other models. The Poisson...Count data that exhibit over dispersion (variance of counts is larger than its mean) are commonly analyzed using discrete distributions such as negative binomial, Poisson inverse Gaussian and other models. The Poisson is characterized by the equality of mean and variance whereas the Negative Binomial and the Poisson inverse Gaussian have variance larger than the mean and therefore are more appropriate to model over-dispersed count data. As an alternative to these two models, we shall use the generalized Poisson distribution for group comparisons in the presence of multiple covariates. This problem is known as the ANCOVA and is solved for continuous data. Our objectives were to develop ANCOVA using the generalized Poisson distribution, and compare its goodness of fit to that of the nonparametric Generalized Additive Models. We used real life data to show that the model performs quite satisfactorily when compared to the nonparametric Generalized Additive Models.展开更多
In this paper, the frequency of an earthquake occurrence and magnitude relationship has been modeled with generalized linear models for the set of earthquake data of Nepal. A goodness of fit of a statistical model is ...In this paper, the frequency of an earthquake occurrence and magnitude relationship has been modeled with generalized linear models for the set of earthquake data of Nepal. A goodness of fit of a statistical model is applied for generalized linear models and considering the model selection information criterion, Akaike information criterion and Bayesian information criterion, generalized Poisson regression model has been selected as a suitable model for the study. The objective of this study is to determine the parameters (a and b values), estimate the probability of an earthquake occurrence and its return period using a Poisson regression model and compared with the Gutenberg-Richter model. The study suggests that the probabilities of earthquake occurrences and return periods estimated by both the models are relatively close to each other. The return periods from the generalized Poisson regression model are comparatively smaller than the Gutenberg-Richter model.展开更多
In this article, we propose a generalized empirical likelihood inference for the parametric component in semiparametric generalized partially linear models with longitudinal data. Based on the extended score vector, a...In this article, we propose a generalized empirical likelihood inference for the parametric component in semiparametric generalized partially linear models with longitudinal data. Based on the extended score vector, a generalized empirical likelihood ratios function is defined, which integrates the within-cluster?correlation meanwhile avoids direct estimating the nuisance parameters in the correlation matrix. We show that the proposed statistics are asymptotically?Chi-squared under some suitable conditions, and hence it can be used to construct the confidence region of parameters. In addition, the maximum empirical likelihood estimates of parameters and the corresponding asymptotic normality are obtained. Simulation studies demonstrate the performance of the proposed method.展开更多
文摘The growing demand for wireless connectivity has made massive multiple-input multiple-output(MIMO)a cornerstone of modern communication systems.To optimize network performance and resource allocation,an efficient and robust approach is joint device activity detection and channel estimation.In this paper,we present an approach utilizing score-based generative models to address the underdetermined nature of channel estimation,which is data-driven and well-suited for the complex and dynamic environment of massive MIMO systems.Our experimental results,based on a comprehensive dataset generated through Monte-Carlo sampling,demonstrate the high precision of our channel estimation approach,with errors reduced to as low as-45 d B,and exceptional accuracy in detecting active devices.
基金Supported by the Henan Provincial Health Commission,No.232102310145.
文摘BACKGROUND Patients with chronic obstructive pulmonary disease(COPD)frequently experience exacerbations requiring multiple hospitalizations over prolonged disease courses,which predispose them to generalized anxiety disorder(GAD).This comorbidity exacerbates breathing difficulties,activity limitations,and social isolation.While previous studies predominantly employed the GAD 7-item scale for screening,this approach is somewhat subjective.The current literature on predictive models for GAD risk in patients with COPD is limited.AIM To construct and validate a GAD risk prediction model to aid healthcare professionals in preventing the onset of GAD.METHODS This retrospective analysis encompassed patients with COPD treated at our institution from July 2021 to February 2024.The patients were categorized into a modeling(MO)group and a validation(VA)group in a 7:3 ratio on the basis of the occurrence of GAD.Univariate and multivariate logistic regression analyses were utilized to construct the risk prediction model,which was visualized using forest plots.The model’s performance was evaluated using Hosmer-Lemeshow(H-L)goodness-of-fit test and receiver operating characteristic(ROC)curve analysis.RESULTS A total of 271 subjects were included,with 190 in the MO group and 81 in the VA group.GAD was identified in 67 patients with COPD,resulting in a prevalence rate of 24.72%(67/271),with 49 cases(18.08%)in the MO group and 18 cases(22.22%)in the VA group.Significant differences were observed between patients with and without GAD in terms of educational level,average household income,smoking history,smoking index,number of exacerbations in the past year,cardiovascular comorbidities,disease knowledge,and personality traits(P<0.05).Multivariate logistic regression analysis revealed that lower education levels,household income<3000 China yuan,smoking history,smoking index≥400 cigarettes/year,≥two exacerbations in the past year,cardiovascular comorbidities,complete lack of disease information,and introverted personality were significant risk factors for GAD in the MO group(P<0.05).ROC analysis indicated that the area under the curve for predicting GAD in the MO and VA groups was 0.978 and 0.960.The H-L test yieldedχ^(2) values of 6.511 and 5.179,with P=0.275 and 0.274.Calibration curves demonstrated good agreement between predicted and actual GAD occurrence risks.CONCLUSION The developed predictive model includes eight independent risk factors:Educational level,household income,smoking history,smoking index,number of exacerbations in the past year,presence of cardiovascular comorbidities,level of disease knowledge,and personality traits.This model effectively predicts the onset of GAD in patients with COPD,enabling early identification of high-risk individuals and providing a basis for early preventive interventions by nursing staff.
文摘Panoramic images, offering a 360-degree view, are essential in virtual reality(VR) and augmented reality(AR), enhancing realism with high-quality textures. However, acquiring complete and high-quality panoramic textures is challenging. This paper introduces a method using generative adversarial networks(GANs) and the contrastive language-image pretraining(CLIP) model to restore and control texture in panoramic images. The GAN model captures complex structures and maintains consistency, while CLIP enables fine-grained texture control via semantic text-image associations. GAN inversion optimizes latent codes for precise texture details. The resulting low dynamic range(LDR) images are converted to high dynamic range(HDR) using the Blender engine for seamless texture blending. Experimental results demonstrate the effectiveness and flexibility of this method in panoramic texture restoration and generation.
基金supported by National Natural Science Foundation of China(62376219 and 62006194)Foundational Research Project in Specialized Discipline(Grant No.G2024WD0146)Faculty Construction Project(Grant No.24GH0201148).
文摘Large language models(LLMs)have undergone significant expansion and have been increasingly integrated across various domains.Notably,in the realm of robot task planning,LLMs harness their advanced reasoning and language comprehension capabilities to formulate precise and efficient action plans based on natural language instructions.However,for embodied tasks,where robots interact with complex environments,textonly LLMs often face challenges due to a lack of compatibility with robotic visual perception.This study provides a comprehensive overview of the emerging integration of LLMs and multimodal LLMs into various robotic tasks.Additionally,we propose a framework that utilizes multimodal GPT-4V to enhance embodied task planning through the combination of natural language instructions and robot visual perceptions.Our results,based on diverse datasets,indicate that GPT-4V effectively enhances robot performance in embodied tasks.This extensive survey and evaluation of LLMs and multimodal LLMs across a variety of robotic tasks enriches the understanding of LLM-centric embodied intelligence and provides forward-looking insights towards bridging the gap in Human-Robot-Environment interaction.
文摘This study explores a novel educational model of generative AI-empowered interdisciplinary project-based learning(PBL).By analyzing the current applications of generative AI technology in information technology curricula,it elucidates its advantages and operational mechanisms in interdisciplinary PBL.Combining case studies and empirical research,the investigation proposes implementation pathways and strategies for the generative AI-enhanced interdisciplinary PBL model,detailing specific applications across three phases:project preparation,implementation,and evaluation.The research demonstrates that generative AI-enabled interdisciplinary project-based learning can effectively enhance students’learning motivation,interdisciplinary thinking capabilities,and innovative competencies,providing new conceptual frameworks and practical approaches for educational model innovation.
文摘This critical review provides an in-depth analysis of Large Language Models(LLMs),encompassing their foundational principles,diverse applications,and advanced training methodologies.We critically examine the evolution from Recurrent Neural Networks(RNNs)to Transformer models,highlighting the significant advancements and innovations in LLM architectures.The review explores state-of-the-art techniques such as in-context learning and various fine-tuning approaches,with an emphasis on optimizing parameter efficiency.We also discuss methods for aligning LLMs with human preferences,including reinforcement learning frameworks and human feedback mechanisms.The emerging technique of retrieval-augmented generation,which integrates external knowledge into LLMs,is also evaluated.Additionally,we address the ethical considerations of deploying LLMs,stressing the importance of responsible and mindful application.By identifying current gaps and suggesting future research directions,this review provides a comprehensive and critical overview of the present state and potential advancements in LLMs.This work serves as an insightful guide for researchers and practitioners in artificial intelligence,offering a unified perspective on the strengths,limitations,and future prospects of LLMs.
基金supported by the National Multidisciplinary Innovation Team of Traditional Chinese Medicine(ZYYCXTD-D-202204)China Postdoctoral Science Foundation(2023M742627)+1 种基金Postdoctoral Fellowship Program of CPSF(GZC20231928)Foundation of State Key Laboratory of Component-based Chinese Medicine(CBCM2023201).
文摘Objective:Generative artificial intelligence(AI)technology,represented by large language models(LLMs),has gradually been developed for traditional Chinese medicine(TCM);however,challenges remain in effectively enhancing AI applications for TCM.Therefore,this study is the first systematic review to analyze LLMs in TCM retrospectively,focusing on and summarizing the evidence of their performance in generative tasks.Methods:We extensively searched electronic databases for articles published until June 2024 to identify publicly available studies on LLMs in TCM.Two investigators independently selected and extracted the related information and evaluation metrics.Based on the available data,this study used descriptive analysis for a comprehensive systematic review of LLM technology related to TCM.Results:Ten studies published between 2023 and 2024 met our eligibility criteria and were included in this review,including 40%LLMs in the TCM vertical domain,40%containing TCM data,and 20%honoring the TCM contribution,with a foundational model parameter range from 1.8 to 33 billion.All included studies used manual or automatic evaluation metrics to evaluate model performance and fully discussed the challenges and contributions through an overview of LLMs in TCM.Conclusions:LLMs have achieved significant advantages in TCM applications and can effectively address intelligent TCM tasks.Further in-depth development of LLMs is needed in various vertical TCM fields,including clinical and fundamental research.Focusing on the functional segmentation development direction of generative AI technologies in TCM application scenarios to meet the practical needs-oriented demands of TCM digitalization is essential.
基金The National Natural Science Foundation of China(No.11171065)the Natural Science Foundation of Jiangsu Province(No.BK2011058)
文摘In order to detect whether the data conforms to the given model, it is necessary to diagnose the data in the statistical way. The diagnostic problem in generalized nonlinear models based on the maximum Lq-likelihood estimation is considered. Three diagnostic statistics are used to detect whether the outliers exist in the data set. Simulation results show that when the sample size is small, the values of diagnostic statistics based on the maximum Lq-likelihood estimation are greater than the values based on the maximum likelihood estimation. As the sample size increases, the difference between the values of the diagnostic statistics based on two estimation methods diminishes gradually. It means that the outliers can be distinguished easier through the maximum Lq-likelihood method than those through the maximum likelihood estimation method.
基金supported by MedTechLabs,GE HealthCare,the Swedish Research council,No.2021-05103the Göran Gustafsson foundation,No.2114.
文摘Deep learning(DL)has proven to be important for computed tomography(CT)image denoising.However,such models are usually trained under supervision,requiring paired data that may be difficult to obtain in practice.Diffusion models offer unsupervised means of solving a wide range of inverse problems via posterior sampling.In particular,using the estimated unconditional score function of the prior distribution,obtained via unsupervised learning,one can sample from the desired posterior via hijacking and regularization.However,due to the iterative solvers used,the number of function evaluations(NFE)required may be orders of magnitudes larger than for single-step samplers.In this paper,we present a novel image denoising technique for photon-counting CT by extending the unsupervised approach to inverse problem solving to the case of Poisson flow generative models(PFGM)++.By hijacking and regularizing the sampling process we obtain a single-step sampler,that is NFE=1.Our proposed method incorporates posterior sampling using diffusion models as a special case.We demonstrate that the added robustness afforded by the PFGM++framework yields significant performance gains.Our results indicate competitive performance compared to popular supervised,including state-of-the-art diffusion-style models with NFE=1(consistency models),unsupervised,and non-DL-based image denoising techniques,on clinical low-dose CT data and clinical images from a prototype photon-counting CT system developed by GE HealthCare.
基金supported by the Central Public Welfare Fund of China (Grant No.Y710005)the China Scholarship Council (CSC)
文摘Soil undergoes both elastic and plastic deformations under different loading conditions. A relatively accurate constitutive model of soil behaviors should be capable of predicting the elastic and plastic deformations properly. Among a large number of elastoplastic constitutive models developed over the last several decades, constitutive models based on generalized plasticity have been successfully utilized in modeling the mechanical behavior of various soils. This paper attempts to present a review of the most recent developments of generalized plasticity models for geotechnical problems. After a brief review of generalized plasticity theories and constitutive models, limitations of the original Pastor-Zienkiewicz model in practical application are summarized. Afterwards, recent achievements in the generalized plasticity models for both saturated and unsaturated soils and their applicability are analyzed, and a general approach for modification of generalized plasticity models is highlighted.
基金Under the auspices of National Natural Science Foundation of China(No.41001363)
文摘This study aims to provide a predictive vegetation mapping approach based on the spectral data, DEM and Generalized Additive Models (GAMs). GAMs were used as a prediction tool to describe the relationship between vegetation and environmental variables, as well as spectral variables. Based on the fitted GAMs model, probability map of species occurrence was generated and then vegetation type of each grid was defined according to the probability of species occurrence. Deviance analysis was employed to test the goodness of curve fitting and drop contribution calculation was used to evaluate the contribution of each predictor in the fitted GAMs models. Area under curve (AUC) of Receiver Operating Characteristic (ROC) curve was employed to assess the results maps of probability. The results showed that: 1) AUC values of the fitted GAMs models are very high which proves that integrating spectral data and environmental variables based on the GAMs is a feasible way to map the vegetation. 2) Prediction accuracy varies with plant community, and community with dense cover is better predicted than sparse plant community. 3) Both spectral variables and environmental variables play an important role in mapping the vegetation. However, the contribution of the same predictor in the GAMs models for different plant communities is different. 4) Insufficient resolution of spectral data, environmental data and confounding effects of land use and other variables which are not closely related to the environmental conditions are the major causes of imprecision.
基金supported by the special research fund of the Ocean University of China (No.201022001)
文摘The catch and effort data of Sillago sihama fishery in Pakistani waters were used to investigate the performance of two closely related stock assessment models: logistic and generalized surplus-production models. Compared with the generalized production model, the logistic model produced more reasonable estimates for parameters such as maximum sustainable yield. The Akaike's Information Criterion values estimated at 4.265 and -51.152 respectively by the logistic and generalized models. Simulation analyses of the S. sihama fishery showed that the estimated and observed abundance indices for the logistic model were closer than those for the generalized production model. Standardized residuals were distributed closer for logistic model, but exhibited a slightly increasing trend for the generalized model. Statistical outliers were seen in 1989 and 1993 for the logistic model, and in 1981 and 1999 for the generalized model. Simulated results revealed that the logistic estimates were close to the true value for low CVs (coefficients of variation) but widely dispersed for high CVs. In contrast, the generalized model estimates were loose for all CV levels. The estimated production model curve parameter was not reasonable at all the tested levels of white noise. With the increase in white noise R2 for the catch per unit effort decreased. Therefore, we conclude that the logistic model performs more reasonably than the generalized production model.
文摘Objective: To analyze longitudinal binary data by using generalized linear models. The correlation between repeated measures were considered. The general method for analyzing longitudinal binary data was given. Methods: Generalized estimating equations (GEE) proposed by Zeger and Liang was used. For sevens covariance structures, one method was given for estimating regression and correlation parameters. Results: Regression and coerelation parameters were estimated simultaneously. A Set of program was finished and an example was illustrated. Conclusion: Longitudinal dsta often occur in medical researches and clinical trials. For solving the problem of correlation between repeated measures, it is necessary to use some special methods to cope with this Kind of data.
基金Under the auspices of the Project of National Natural Science Foundation of China ( No. 41001363)Autonomous Project of State Key Laboratory of Resources and Environmental Information System,Geo-information Tupu Theory and Virtual Geoscience
文摘There are typical ecosystems of littoral wetlands in the Yellow River Delta.In order to study the relationships between Tamarix chinensis and environmental variables and to predict T.chinensis potential distribution in the Yellow River Delta,641 vegetation samples and 964 soil samples were collected in the area in October of 2004,2005,2006 and 2007.The contents of soil organic matter,total phosphorus,salt,and soluble potassium were determined.Then,the analyzed data were interpolated into spatial raster data by Kriging interpolation method.Meanwhile,the digital elevation model,soil type map and landform unit map of the Yellow River Delta were also collected.Generalized Additive Models(GAMs) were employed to build species-environment model and then simulate the potential distribution of T.chinensis.The results indicated that the distribution of T.chinensis was mainly limited by soil salt content,total soil phosphorus content,soluble potassium content,soil type,landform unit,and elevation.The distribution probability of T.chinensis was produced with a lookup table generated by Grasp Module(based on GAMs) in software ArcView GIS 3.2.The AUC(Area Under Curve) value of validation and cross-validation of ROC(Receive Operating Characteristic) were both higher than 0.8,which suggested that the established model had a high precision for predicting species distribution.
文摘On foe basis of the Kirchoff-Karman hypothses for the nonlinear bending of thin plates, the three kinds of boundary value problems of nonlinear analysis for perforated fhin plates are presented under the differenr in-plane boundary conditions and the corresponding generalized varialional principles are established. One can see that all mathematical models presented in this paper are completely new ones and differ from the ordinary von Karman theory. These mathematical models can be applied to the nonlinear analysis and the Stability analysis of perforaled thin plates in arbitraryplane boundary conditions.
基金Supported by the National Natural Science Foundation of China （61273131） 111 Project （B12018）+1 种基金 the Innovation Project of Graduate in Jiangsu Province （CXZZ12_0741） the Fundamental Research Funds for the Central Universities （JUDCF12034）
文摘Fault monitoring of bioprocess is important to ensure safety of a reactor and maintain high quality of products. It is difficult to build an accurate mechanistic model for a bioprocess, so fault monitoring based on rich historical or online database is an effective way. A group of data based on bootstrap method could be resampling stochastically, improving generalization capability of model. In this paper, online fault monitoring of generalized additive models （GAMs） combining with bootstrap is proposed for glutamate fermentation process. GAMs and bootstrap are first used to decide confidence interval based on the online and off-line normal sampled data from glutamate fermentation experiments. Then GAMs are used to online fault monitoring for time, dissolved oxygen, oxygen uptake rate, and carbon dioxide evolution rate. The method can provide accurate fault alarm online and is helpful to provide useful information for removing fault and abnormal phenomena in the fermentation.
文摘We study the quasi likelihood equation in Generalized Linear Models（GLM） with adaptive design ∑（i=1）^n xi（yi-h（x＇iβ））=0, where yi is a q=vector, and xi is a p×q random matrix. Under some assumptions, it is shown that the Quasi- Likelihood equation for the GLM has a solution which is asymptotic normal.
文摘Count data that exhibit over dispersion (variance of counts is larger than its mean) are commonly analyzed using discrete distributions such as negative binomial, Poisson inverse Gaussian and other models. The Poisson is characterized by the equality of mean and variance whereas the Negative Binomial and the Poisson inverse Gaussian have variance larger than the mean and therefore are more appropriate to model over-dispersed count data. As an alternative to these two models, we shall use the generalized Poisson distribution for group comparisons in the presence of multiple covariates. This problem is known as the ANCOVA and is solved for continuous data. Our objectives were to develop ANCOVA using the generalized Poisson distribution, and compare its goodness of fit to that of the nonparametric Generalized Additive Models. We used real life data to show that the model performs quite satisfactorily when compared to the nonparametric Generalized Additive Models.
文摘In this paper, the frequency of an earthquake occurrence and magnitude relationship has been modeled with generalized linear models for the set of earthquake data of Nepal. A goodness of fit of a statistical model is applied for generalized linear models and considering the model selection information criterion, Akaike information criterion and Bayesian information criterion, generalized Poisson regression model has been selected as a suitable model for the study. The objective of this study is to determine the parameters (a and b values), estimate the probability of an earthquake occurrence and its return period using a Poisson regression model and compared with the Gutenberg-Richter model. The study suggests that the probabilities of earthquake occurrences and return periods estimated by both the models are relatively close to each other. The return periods from the generalized Poisson regression model are comparatively smaller than the Gutenberg-Richter model.
文摘In this article, we propose a generalized empirical likelihood inference for the parametric component in semiparametric generalized partially linear models with longitudinal data. Based on the extended score vector, a generalized empirical likelihood ratios function is defined, which integrates the within-cluster?correlation meanwhile avoids direct estimating the nuisance parameters in the correlation matrix. We show that the proposed statistics are asymptotically?Chi-squared under some suitable conditions, and hence it can be used to construct the confidence region of parameters. In addition, the maximum empirical likelihood estimates of parameters and the corresponding asymptotic normality are obtained. Simulation studies demonstrate the performance of the proposed method.
