Evaluating Privacy Leakage and Memorization Attacks on Large Language Models (LLMs) in Generative AI Applications

The recent interest in the deployment of Generative AI applications that use large language models (LLMs) has brought to the forefront significant privacy concerns, notably the leakage of Personally Identifiable Information (PII) and other confidential or protected information that may have been memorized during training, specifically during a fine-tuning or customization process. We describe different black-box attacks from potential adversaries and study their impact on the amount and type of information that may be recovered from commonly used and deployed LLMs. Our research investigates the relationship between PII leakage, memorization, and factors such as model size, architecture, and the nature of attacks employed. The study utilizes two broad categories of attacks: PII leakage-focused attacks (auto-completion and extraction attacks) and memorization-focused attacks (various membership inference attacks). The findings from these investigations are quantified using an array of evaluative metrics, providing a detailed understanding of LLM vulnerabilities and the effectiveness of different attacks.

Research status and application of artificial intelligence large models in the oil and gas industry

This article elucidates the concept of large model technology,summarizes the research status of large model technology both domestically and internationally,provides an overview of the application status of large models in vertical industries,outlines the challenges and issues confronted in applying large models in the oil and gas sector,and offers prospects for the application of large models in the oil and gas industry.The existing large models can be briefly divided into three categories:large language models,visual large models,and multimodal large models.The application of large models in the oil and gas industry is still in its infancy.Based on open-source large language models,some oil and gas enterprises have released large language model products using methods like fine-tuning and retrieval augmented generation.Scholars have attempted to develop scenario-specific models for oil and gas operations by using visual/multimodal foundation models.A few researchers have constructed pre-trained foundation models for seismic data processing and interpretation,as well as core analysis.The application of large models in the oil and gas industry faces challenges such as current data quantity and quality being difficult to support the training of large models,high research and development costs,and poor algorithm autonomy and control.The application of large models should be guided by the needs of oil and gas business,taking the application of large models as an opportunity to improve data lifecycle management,enhance data governance capabilities,promote the construction of computing power,strengthen the construction of“artificial intelligence+energy”composite teams,and boost the autonomy and control of large model technology.

On the Application of Mixed Models of Probability and Convex Set for Time-Variant Reliability Analysis

In time-variant reliability problems,there are a lot of uncertain variables from different sources.Therefore,it is important to consider these uncertainties in engineering.In addition,time-variant reliability problems typically involve a complexmultilevel nested optimization problem,which can result in an enormous amount of computation.To this end,this paper studies the time-variant reliability evaluation of structures with stochastic and bounded uncertainties using a mixed probability and convex set model.In this method,the stochastic process of a limit-state function with mixed uncertain parameters is first discretized and then converted into a timeindependent reliability problem.Further,to solve the double nested optimization problem in hybrid reliability calculation,an efficient iterative scheme is designed in standard uncertainty space to determine the most probable point(MPP).The limit state function is linearized at these points,and an innovative random variable is defined to solve the equivalent static reliability analysis model.The effectiveness of the proposed method is verified by two benchmark numerical examples and a practical engineering problem.

Factors affecting farmers'choice to adopt risk management strategies:The application of multivariate and multinomial probit models

This study investigates the factors that impact farmers'adoption of risk management strategies(RMS)in Pakistan during times of uncertainty.The study examines farmers'adoption of RMS using both multinomial probit(MNP)and multivariate probit(MVP).Data were collected from 382 farmers sampled from four districts in KhyberPakhtunkhwa(KP)province of Pakistan via a multistage sampling technique.This study utilizes the MNP model,considering the assumption of Independence of Irrelevant Alternatives(IIA)and incorporating correlated error terms.The objective is to understand farmers'behavior in risky situations and determine if there is heterogeneity.Results are compared with the MVP model to assess robustness and gain deeper understanding of farmers'decisionmaking processes.The research findings reveal that our results are robust,and farmers behave homogeneously in various RMS scenarios.Farmers adopt RMS individually or in combination to mitigate the adverse effects of natural calamities on their livelihood.The risk-averse farmers,who perceive weather-related risks as a threat,access credits and information,and have farms close to a river are more likely to adopt RMS,irrespective of the format of the strategies available.Moreover,the predicted probabilities and correlation of the RMS and RM categories have strengthened our model estimation.These findings provide insights into the behavior of farmers in adopting RMS which are helpful for policymakers and stakeholders in developing strategies to mitigate the impacts of natural calamities on farmers.

Analysis of application range of simplified models for field to thermo-field to thermionic emission processes from the cathode

Electron emission plays a dominant role in plasma-cathode interactions and is a key factor in many plasma phenomena and industrial applications.It is necessary to illustrate the various electron emission mechanisms and the corresponding applicable description models to evaluate their impacts on discharge properties.In this study,detailed expressions of the simplified formulas valid for field emission to thermo-field emission to thermionic emission typically used in the numerical simulation are proposed,and the corresponding application ranges are determined in the framework of the Murphy-Good theory,which is commonly regarded as the general model and to be accurate in the full range of conditions of the validity of the theory.Dimensionless parameterization was used to evaluate the emission current density of the Murphy-Good formula,and a deviation factor was defined to obtain the application ranges for different work functions(2.5‒5 eV),cathode temperatures(300‒6000 K),and emitted electric fields(10^(5) to 10^(10) V·m^(-1)).The deviation factor was shown to be a nonmonotonic function of the three parameters.A comparative study of particle number densities in atmospheric gas discharge with a tungsten cathode was performed based on the one-dimensional implicit particle-in-cell(PIC)with the Monte Carlo collision(MCC)method according to the aforementioned application ranges.It was found that small differences in emission current density can lead to variations in the distributions of particle number density due to changes in the collisional environment.This study provides a theoretical basis for selecting emission models for subsequent numerical simulations.

Evaluation and application of kinetic models for Cu-catalyzed acetylene hydrochlorination

The development of environmentally friendly catalysts has become a top priority for acetylene hydrochlorination.However,difficulties remain in systematic studies on the applicability of kinetic models for the industrialization of Cu-based catalysts.Therefore,a strategy involving reactor modeling,parameter estimation,and model testing is developed to evaluate the predictive ability of kinetic models.In order to search for reliable and widely applicable reaction kinetic models for Cu-based catalysts,a case study is conducted.Multiple possible kinetic models derived from the power law,adsorption mechanism,and reaction path are sifted through collecting and testing activity data from tens of Cu-based catalysts.Different optimum applicable ranges of these kinetic models are presented.According to the comparative analysis on their applications in various industrial scenarios,this research suggests that kinetic models derived from reaction path exhibits the best extrapolation ability and has the greatest potential for application in the scale-up design of reactors.

Applications and advancements in animal models for antiviral research on mosquito-borne arboviruses

Vector-borne diseases caused by arthropod-borne viruses(arboviruses) are a considerable challenge to public health globally. Mosquito-borne arboviruses, such as Chikungunya, Dengue, and Zika viruses, cause a range of human illnesses and may be fatal. Currently, efforts to control these diseases still face challenges due to growing vector resistance towards insecticides, urbanization, and limited effective antiviral treatments and vaccines. Animal models are crucial in antiviral research on mosquito-borne arboviruses, playing a role in understanding disease mechanisms,vaccine development, and toxicity testing, but the application of animal models still faces the challenges of ethical considerations and animal-to-human translational success. Genetically engineered mouse models, hamster models and non-human primate(NHP) are currently used in arbovirus research, but new models such as tree shrews and novel humanized mice are emerging. In the context of Malaysian research, the use of long-tailed macaques as potential NHP models for arbovirus research is possible;however, it faces the ethical dilemma of using an endangered species for scientific purposes. Overall, animal models play a crucial role in advancing infectious disease research, but a balance between medical research and species conservation must be upheld.

Advances in memristor based artificial neuron fabrication-materials,models,and applications

Spiking neural network(SNN),widely known as the third-generation neural network,has been frequently investigated due to its excellent spatiotemporal information processing capability,high biological plausibility,and low energy consumption characteristics.Analogous to the working mechanism of human brain,the SNN system transmits information through the spiking action of neurons.Therefore,artificial neurons are critical building blocks for constructing SNN in hardware.Memristors are drawing growing attention due to low consumption,high speed,and nonlinearity characteristics,which are recently introduced to mimic the functions of biological neurons.Researchers have proposed multifarious memristive materials including organic materials,inorganic materials,or even two-dimensional materials.Taking advantage of the unique electrical behavior of these materials,several neuron models are successfully implemented,such as Hodgkin–Huxley model,leaky integrate-and-fire model and integrate-and-fire model.In this review,the recent reports of artificial neurons based on memristive devices are discussed.In addition,we highlight the models and applications through combining artificial neuronal devices with sensors or other electronic devices.Finally,the future challenges and outlooks of memristor-based artificial neurons are discussed,and the development of hardware implementation of brain-like intelligence system based on SNN is also prospected.

Nonlinear constitutive models of rock structural plane and their applications

Structural planes play an important role in controlling the stability of rock engineering,and the influence of structural planes should be considered in the design and construction process of rock engineering.In this paper,mechanical properties,constitutive theory,and numerical application of structural plane are studied by a combination method of laboratory tests,theoretical derivation,and program development.The test results reveal the change laws of various mechanical parameters under different roughness and normal stress.At the pre-peak stage,a non-stationary model of shear stiffness is established,and threedimensional empirical prediction models for initial shear stiffness and residual stage roughness are proposed.The nonlinear constitutive models are established based on elasto-plastic mechanics,and the algorithms of the models are developed based on the return mapping algorithm.According to a large number of statistical analysis results,empirical prediction models are proposed for model parameters expressed by structural plane characteristic parameters.Finally,the discrete element method(DEM)is chosen to embed the constitutive models for practical application.The running programs of the constitutive models have been compiled into the discrete element model library.The comparison results between the proposed model and the Mohr-Coulomb slip model show that the proposed model can better describe nonlinear changes at different stages,and the predicted shear strength,peak strain and shear stiffness are closer to the test results.The research results of the paper are conducive to the accurate evaluation of structural plane in rock engineering.

Application of Patient Simulators Combined with Internet plus Scenario Simulation Teaching Models on Intravenous Infusion Nursing Education in China

Objective: To explore the effectiveness of applying patient simulators combined with Internet Plus scenario simulation teaching models on intravenous (IV) infusion nursing education, and to provide scientific evidence for the implementation of advanced teaching models in future nursing education. Methods: Enrolled 60 nurses who took the IV infusion therapy training program in our hospital from January 2022 to December 2023 for research. 30 nurses who were trained in traditional teaching models from January to December 2022 were selected as the control group, and 30 nurses who were trained with simulation-based teaching models with methods including simulated patients, internet, online meetings which can be replayed and scenario simulation, etc. from January to December 2023 were selected as the experimental group. Evaluated the learning outcomes based on the Competency Inventory for Nursing Students (CINS), Problem-Solving Inventory (PSI), comprehensive learning ability, scientific research ability, and proficiency in the theoretical knowledge and practical skills of IV infusion therapy. Nursing quality, the incidence of IV infusion therapy complications and nurse satisfaction with different teaching models were also measured. Results: The scientific research ability, PSI scores, CINS scores, and comprehensive learning ability of the experimental group were better than those of the control group (P 0.05), and their assessment results of practical skills, nursing quality of IV infusion therapy during training, and satisfaction with teaching models were all better than those of the control group with statistical significance (P < 0.05). The incidence of IV infusion therapy complications in the experimental group was lower than that in the control group, indicating statistical significance (P < 0.05). Conclusions: Teaching models based on patient simulators combined with Internet Plus scenario simulation enable nursing students to learn more directly and practice at any time and in any place, and can improve their proficiency in IV infusion theoretical knowledge and skills (e.g. PICC catheterization), core competencies, problem-solving ability, comprehensive learning ability, scientific research ability and the ability to deal with complicated cases. Also, it helps provide high-quality nursing education, improve the nursing quality of IV therapy, reduce the incidence of related complications, and ensure the safety of patients with IV therapy.

Comparative Analysis of the Factors Influencing Metro Passenger Arrival Volumes in Wuhan, China, and Lagos, Nigeria: An Application of Association Rule Mining and Neural Network Models

This study explores the factors influencing metro passengers’ arrival volume in Wuhan, China, and Lagos, Nigeria, by examining weather, time of day, waiting time, travel behavior, arrival patterns, and metro satisfaction. It addresses a significant research gap in understanding metro passengers’ dynamics across cultural and geographical contexts. It employs questionnaires, field observations, and advanced data analysis techniques like association rule mining and neural network modeling. Key findings include a correlation between rainy weather, shorter waiting times, and higher arrival volumes. Neural network models showed high predictive accuracy, with waiting time, metro satisfaction, and weather being significant factors in Lagos Light Rail Blue Line Metro. In contrast, arrival patterns, weather, and time of day were more influential in Wuhan Metro Line 5. Results suggest that improving metro satisfaction and reducing waiting times could increase arrival volumes in Lagos Metro while adjusting schedules for weather and peak times could optimize flow in Wuhan Metro. These insights are valuable for transportation planning, passenger arrival volume management, and enhancing user experiences, potentially benefiting urban transportation sustainability and development goals.

Research on the Construction and Application of Service Quality Evaluation Models for Scenic Spots:Taking Chongqing Ciqikou as an Example

In the current environment of increasingly fierce competition in the tourism industry,service quality has become crucial for enhancing the competitiveness of scenic spots.This paper uses the SERVQUAL model to design a service quality evaluation questionnaire that captures the gap between tourists’expectations and perceptions,using the Ciqikou Scenic Spot as a case study.Data collected from field surveys are used to comprehensively and meticulously evaluate the service quality of the Ciqikou Scenic Spot.The analysis results show that the scenic spot,with its unique folk culture experience and beautiful ecological environment,has certain advantages in terms of service quality.However,significant deficiencies exist in infrastructure and environmental hygiene.Accordingly,targeted improvement suggestions are proposed to further enhance the service quality of the Ciqikou Scenic Spot and meet the increasingly diverse and personalized needs of tourists.This study provides not only a specific service quality improvement strategy for the Ciqikou Scenic Spot but also a valuable reference for other tourist attractions.

The evolution of robotics:research and application progress of dental implant robotic systems

The use of robots to augment human capabilities and assist in work has long been an aspiration.Robotics has been developing since the 1960s when the first industrial robot was introduced.As technology has advanced,robotic-assisted surgery has shown numerous advantages,including more precision,efficiency,minimal invasiveness,and safety than is possible with conventional techniques,which are research hotspots and cutting-edge trends.This article reviewed the history of medical robot development and seminal research papers about current research progress.Taking the autonomous dental implant robotic system as an example,the advantages and prospects of medical robotic systems would be discussed which would provide a reference for future research.

Novel models for simulating maize growth based on thermal time and photothermal units: Applications under various mulching practices

Maize (Zea mays L.) is one of the three major food crops and an important source of carbohydrates for maintaining food security around the world.Plant height (H),stem diameter (SD),leaf area index (LAI) and dry matter (DM) are important growth parameters that influence maize production.However,the combined effect of temperature and light on maize growth is rarely considered in crop growth models.Ten maize growth models based on the modified logistic growth equation (Mlog) and the Mitscherlich growth equation (Mit) were proposed to simulate the H,SD,LAI and DM of maize under different mulching practices based on experimental data from 2015–2018.Either the accumulative growing degree-days (AGDD),helio thermal units (HTU),photothermal units (PTU) or photoperiod thermal units (PPTU,first proposed here) was used as a single driving factor in the models;or AGDD was combined with either accumulative actual solar hours (ASS),accumulative photoperiod response (APR,first proposed here) or accumulative maximum possible sunshine hours (ADL) as the dual driving factors in the models.The model performances were evaluated using seven statistical indicators and a global performance index.The results showed that the three mulching practices significantly increased the maize growth rates and the maximum values of the growth curves compared with non-mulching.Among the four single factor-driven models,the overall performance of the Mlog_(PTU)Model was the best,followed by the Mlog_(AGDD)Model.The Mlog_(PPTU)Model was better than the Mlog_(AGDD)Model in simulating SD and LAI.Among the 10 models,the overall performance of the Mlog_(AGDD–APR)Model was the best,followed by the Mlog_(AGDD–ASS)Model.Specifically,the Mlog_(AGDD–APR)Model performed the best in simulating H and LAI,while the Mlog_(AGDD–ADL)and Mlog_(AGDD–ASS)models performed the best in simulating SD and DM,respectively.In conclusion,the modified logistic growth equations with AGDD and either APR,ASS or ADL as the dual driving factors outperformed the commonly used modified logistic growth model with AGDD as a single driving factor in simulating maize growth.

Potential industrial applications of photo/electrocatalysis: Recent progress and future challenges

Nowadays,the rapid development of the social economy inevitably leads to global energy and environmental crisis.For this reason,more and more scholars focus on the development of photocatalysis and/or electrocatalysis technology for the advantage in the sustainable production of high-value-added products,and the high efficiency in pollutants remediation.Although there is plenty of outstanding research has been put forward continuously,most of them focuses on catalysis performance and reaction mechanisms in laboratory conditions.Realizing industrial application of photo/electrocatalytic processes is still a challenge that needs to be overcome by social demand.In this regard,this review comprehensively summarized several explorations in thefield of photo/electrocatalytic reduction towards potential industrial applications in recent years.Special attention is paid to the successful attempts and the current status of photo/electrocatalytic water splitting,carbon dioxide conversion,resource utilization from waste,etc.,by using advanced reactors.The key problems and challenges of photo/electrocatalysis in future industrial practice are also discussed,and the possible development directions are also pointed out from the industry view.

Drilling-based measuring method for the c-φ parameter of rock and its field application

The technology of drilling tests makes it possible to obtain the strength parameter of rock accurately in situ. In this paper, a new rock cutting analysis model that considers the influence of the rock crushing zone(RCZ) is built. The formula for an ultimate cutting force is established based on the limit equilibrium principle. The relationship between digital drilling parameters(DDP) and the c-φ parameter(DDP-cφ formula, where c refers to the cohesion and φ refers to the internal friction angle) is derived, and the response of drilling parameters and cutting ratio to the strength parameters is analyzed. The drillingbased measuring method for the c-φ parameter of rock is constructed. The laboratory verification test is then completed, and the difference in results between the drilling test and the compression test is less than 6%. On this basis, in-situ rock drilling tests in a traffic tunnel and a coal mine roadway are carried out, and the strength parameters of the surrounding rock are effectively tested. The average difference ratio of the results is less than 11%, which verifies the effectiveness of the proposed method for obtaining the strength parameters based on digital drilling. This study provides methodological support for field testing of rock strength parameters.

Anisotropic time-dependent behaviors of shale under direct shearing and associated empirical creep models

Understanding the anisotropic creep behaviors of shale under direct shearing is a challenging issue.In this context,we conducted shear-creep and steady-creep tests on shale with five bedding orientations (i.e.0°,30°,45°,60°,and 90°),under multiple levels of direct shearing for the first time.The results show that the anisotropic creep of shale exhibits a significant stress-dependent behavior.Under a low shear stress,the creep compliance of shale increases linearly with the logarithm of time at all bedding orientations,and the increase depends on the bedding orientation and creep time.Under high shear stress conditions,the creep compliance of shale is minimal when the bedding orientation is 0°,and the steady-creep rate of shale increases significantly with increasing bedding orientations of 30°,45°,60°,and 90°.The stress-strain values corresponding to the inception of the accelerated creep stage show an increasing and then decreasing trend with the bedding orientation.A semilogarithmic model that could reflect the stress dependence of the steady-creep rate while considering the hardening and damage process is proposed.The model minimizes the deviation of the calculated steady-state creep rate from the observed value and reveals the behavior of the bedding orientation's influence on the steady-creep rate.The applicability of the five classical empirical creep models is quantitatively evaluated.It shows that the logarithmic model can well explain the experimental creep strain and creep rate,and it can accurately predict long-term shear creep deformation.Based on an improved logarithmic model,the variations in creep parameters with shear stress and bedding orientations are discussed.With abovementioned findings,a mathematical method for constructing an anisotropic shear creep model of shale is proposed,which can characterize the nonlinear dependence of the anisotropic shear creep behavior of shale on the bedding orientation.

Evolution and Prospects of Foundation Models: From Large Language Models to Large Multimodal Models

Since the 1950s,when the Turing Test was introduced,there has been notable progress in machine language intelligence.Language modeling,crucial for AI development,has evolved from statistical to neural models over the last two decades.Recently,transformer-based Pre-trained Language Models(PLM)have excelled in Natural Language Processing(NLP)tasks by leveraging large-scale training corpora.Increasing the scale of these models enhances performance significantly,introducing abilities like context learning that smaller models lack.The advancement in Large Language Models,exemplified by the development of ChatGPT,has made significant impacts both academically and industrially,capturing widespread societal interest.This survey provides an overview of the development and prospects from Large Language Models(LLM)to Large Multimodal Models(LMM).It first discusses the contributions and technological advancements of LLMs in the field of natural language processing,especially in text generation and language understanding.Then,it turns to the discussion of LMMs,which integrates various data modalities such as text,images,and sound,demonstrating advanced capabilities in understanding and generating cross-modal content,paving new pathways for the adaptability and flexibility of AI systems.Finally,the survey highlights the prospects of LMMs in terms of technological development and application potential,while also pointing out challenges in data integration,cross-modal understanding accuracy,providing a comprehensive perspective on the latest developments in this field.

A Review of Contact Electrification at Diversified Interfaces and Related Applications on Triboelectric Nanogenerator

The triboelectric nanogenerator(TENG)can effectively collect energy based on contact electrification(CE)at diverse interfaces,including solid–solid,liquid–solid,liquid–liquid,gas–solid,and gas–liquid.This enables energy harvesting from sources such as water,wind,and sound.In this review,we provide an overview of the coexistence of electron and ion transfer in the CE process.We elucidate the diverse dominant mechanisms observed at different interfaces and emphasize the interconnectedness and complementary nature of interface studies.The review also offers a comprehensive summary of the factors influencing charge transfer and the advancements in interfacial modification techniques.Additionally,we highlight the wide range of applications stemming from the distinctive characteristics of charge transfer at various interfaces.Finally,this review elucidates the future opportunities and challenges that interface CE may encounter.We anticipate that this review can offer valuable insights for future research on interface CE and facilitate the continued development and industrialization of TENG.

MXene-Based Elastomer Mimetic StretchableSensors: Design, Properties, and Applications

Flexible sensors based on MXene-polymer composites are highly prospective for next-generation wearable electronics used in human-machine interfaces.One of the motivating factors behind the progress of flexible sensors is the steady arrival of new conductive materials.MXenes,a new family of 2D nanomaterials,have been draw-ing attention since the last decade due to their high electronic conduc-tivity,processability,mechanical robustness and chemical tunability.In this review,we encompass the fabrication of MXene-based polymeric nanocomposites,their structure-property relationship,and applications in the flexible sensor domain.Moreover,our discussion is not only lim-ited to sensor design,their mechanism,and various modes of sensing platform,but also their future perspective and market throughout the world.With our article,we intend to fortify the bond between flexible matrices and MXenes thus promoting the swift advancement of flexible MXene-sensors for wearable technologies.