OptoGPT: A foundation model for inverse design in optical multilayer thin film structures

Optical multilayer thin film structures have been widely used in numerous photonic applications.However,existing inverse design methods have many drawbacks because they either fail to quickly adapt to different design targets,or are difficult to suit for different types of structures,e.g.,designing for different materials at each layer.These methods also cannot accommodate versatile design situations under different angles and polarizations.In addition,how to benefit practical fabrications and manufacturing has not been extensively considered yet.In this work,we introduce OptoGPT(Opto Generative Pretrained Transformer),a decoder-only transformer,to solve all these drawbacks and issues simultaneously.

Task-Oriented Semantic Communication with Foundation Models

In the future development direction of the sixth generation(6G)mobile communication,several communication models are proposed to face the growing challenges of the task.The rapid development of artificial intelligence(AI)foundation models provides significant support for efficient and intelligent communication interactions.In this paper,we propose an innovative semantic communication paradigm called task-oriented semantic communication system with foundation models.First,we segment the image by using task prompts based on the segment anything model(SAM)and contrastive language-image pretraining(CLIP).Meanwhile,we adopt Bezier curve to enhance the mask to improve the segmentation accuracy.Second,we have differentiated semantic compression and transmission approaches for segmented content.Third,we fuse different semantic information based on the conditional diffusion model to generate high-quality images that satisfy the users'specific task requirements.Finally,the experimental results show that the proposed system compresses the semantic information effectively and improves the robustness of semantic communication.

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.

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.

Material-SAM:Adapting SAM for Material XCT

X-ray Computed Tomography(XCT)enables non-destructive acquisition of the internal structure of materials,and image segmentation plays a crucial role in analyzing material XCT images.This paper proposes an image segmentation method based on the Segment Anything model(SAM).We constructed a dataset of carbide in nickel-based single crystal superalloys XCT images and preprocessed the images using median filtering,histogram equalization,and gamma correction.Subsequently,SAM was fine-tuned to adapt to the task of material XCT image segmentation,resulting in Material-SAM.We compared the performance of threshold segmentation,SAM,U-Net model,and Material-SAM.Our method achieved 88.45%Class Pixel Accuracy(CPA)and 88.77%Dice Similarity Coefficient(DSC)on the test set,outperforming SAM by 5.25%and 8.81%,respectively,and achieving the highest evaluation.Material-SAM demonstrated lower input requirements compared to SAM,as it only required three reference points for completing the segmentation task,which is one-fifth of the requirement of SAM.Material-SAM exhibited promising results,highlighting its potential as a novel method for material XCT image segmentation.

Prediction Model of Abutment Pressure Affected by Far-Field Hard Stratum Based on Elastic Foundation Theory

In view of the three-dimensional dynamic abutment pressure,the influence of the far-field hard stratum(FHS)in deep,thick coal seams is indeterminant.Based on elastic foundation theory,a three-dimensional dynamic prediction model of the abutment pressure was established.Using this model,the dynamic change in the coal seam abutment pressure caused by the movement of the FHS was studied,and a method for determining the dynamic change range of the abutment pressure was developed.The results of the new prediction model of the abutment pressure are slightly higher than the measured values,with an error of 0.51%,which avoids the shortcomings of the results because the Winkler foundation model results are lower than the measured values and have an error of 9.98%.As time progresses,the abutment pressure and its distribution range are affected by the FHS movement,which has the characteristics of gradually increasing dynamic change until the FHS fractures.The peak value of the abutment pressure increases linearly with time,and the influence range increases with time following a power function with an exponent of less than 1.The influence range of the FHS movement on the abutment pressure ahead of the working face,behind the working face,and along the working face is 10 times,25 times,and 17 times the mining thickness,respectively.According to the actual geological parameters,the dynamic change range of the coal seam abutment pressure was determined by drawing an additional stress curve and by determining the threshold value.These research results are of great significance to the partition optimization of the roadway support design of deep,thick coal seams.

Can Digital Intelligence and Cyber-Physical-Social Systems Achieve Global Food Security and Sustainability?

Plants sequester carbon through photosynthesis and provide primary productivity for the ecosystem. However, they also simultaneously consume water through transpiration, leading to a carbon-water balance relationship. Agricultural production can be regarded as a form of carbon sequestration behavior.From the perspective of the natural-social-economic complex ecosystem, excessive water usage in food production will aggravate regional water pressure for both domestic and industrial purposes. Hence, achieving a harmonious equilibrium between carbon and water resources during the food production process is a key scientific challenge for ensuring food security and sustainability. Digital intelligence(DI) and cyber-physical-social systems(CPSS) are emerging as the new research paradigms that are causing a substantial shift in the conventional thinking and methodologies across various scientific fields, including ecological science and sustainability studies. This paper outlines our recent efforts in using advanced technologies such as big data, artificial intelligence(AI), digital twins, metaverses, and parallel intelligence to model, analyze, and manage the intricate dynamics and equilibrium among plants, carbon, and water in arid and semiarid ecosystems. It introduces the concept of the carbon-water balance and explores its management at three levels: the individual plant level, the community level, and the natural-social-economic complex ecosystem level. Additionally, we elucidate the significance of agricultural foundation models as fundamental technologies within this context. A case analysis of water usage shows that, given the limited availability of water resources in the context of the carbon-water balance, regional collaboration and optimized allocation have the potential to enhance the utilization efficiency of water resources in the river basin. A suggested approach is to consider the river basin as a unified entity and coordinate the relationship between the upstream, midstream and downstream areas. Furthermore, establishing mechanisms for water resource transfer and trade among different industries can be instrumental in maximizing the benefits derived from water resources.Finally, we envisage a future of agriculture characterized by the integration of digital, robotic and biological farming techniques.This vision aims to incorporate small tasks, big models, and deep intelligence into the regular ecological practices of intelligent agriculture.

Foundation models meet visualizations: Challenges and opportunities

Recent studies have indicated that foundation models, such as BERT and GPT, excel atadapting to various downstream tasks. This adaptability has made them a dominant force in buildingartificial intelligence (AI) systems. Moreover, a newresearch paradigm has emerged as visualizationtechniques are incorporated into these models. Thisstudy divides these intersections into two researchareas: visualization for foundation model (VIS4FM)and foundation model for visualization (FM4VIS).In terms of VIS4FM, we explore the primary roleof visualizations in understanding, refining, and evaluating these intricate foundation models. VIS4FMaddresses the pressing need for transparency, explainability, fairness, and robustness. Conversely, in termsof FM4VIS, we highlight how foundation models canbe used to advance the visualization field itself. Theintersection of foundation models with visualizations ispromising but also introduces a set of challenges. Byhighlighting these challenges and promising opportunities, this study aims to provide a starting point forthe continued exploration of this research avenue.

Opportunities for large language models and discourse in engineering

In recent years, large language models have achieved breakthroughs on a wide range of benchmarks in natural language processing and continue to increase in performance. Recently, the advances of large language models have raised interest outside the natural language processing community and could have a large impact on daily life. In this paper, we pose the question: How will large language models and other foundation models shape the future product development process? We provide the reader with an overview of the subject by summarizing both recent advances in natural language processing and the use of information technology in the engineering design process. We argue that discourse should be regarded as the core of engineering design processes, and therefore should be represented in a digital artifact. On this basis, we describe how foundation models such as large language models could contribute to the design discourse by automating parts thereof that involve creativity and reasoning, and were previously reserved for humans. We describe how simulations, experiments, topology optimizations, and other process steps can be integrated into a machine-actionable, discourse-centric design process. As an example, we present a design discourse on the optimization of wind turbine blades. Finally, we outline the future research that will be necessary for the implementation of the conceptualized framework.

A reliable knowledge processing framework for combustion science using foundation models

This research explores the integration of large language models (LLMs) into scientific data assimilation, focusing on combustion science as a case study. Leveraging foundational models integrated with Retrieval-Augmented Generation (RAG) framework, the study introduces an approach to process diverse combustion research data, spanning experimental studies, simulations, and literature. The multifaceted nature of combustion research emphasizes the critical role of knowledge processing in navigating and extracting valuable information from a vast and diverse pool of sources. The developed approach minimizes computational and economic expenses while optimizing data privacy and accuracy. It incorporates prompt engineering and offline open-source LLMs, offering user autonomy in selecting base models. The study provides a thorough examination of text segmentation strategies, conducts comparative studies between LLMs, and explores various optimized prompts to demonstrate the effectiveness of the framework. By incorporating an external vector database, the framework outperforms a conventional LLM in generating accurate responses and constructing robust arguments. Additionally, the study delves into the investigation of optimized prompt templates for the purpose of efficient extraction of scientific literature. Furthermore, we present a targeted scaling study to quantify the algorithmic performance of the framework as the number of prompt tokens increases. The research addresses concerns related to hallucinations and false research articles by introducing a custom workflow developed with a detection algorithm to filter out inaccuracies. Despite identified areas for improvement, the framework consistently delivers accurate domain-specific responses with minimal human oversight. The prompt-agnostic approach introduced holds promise for future improvements. The study underscores the significance of integrating LLMs and knowledge processing techniques in scientific research, providing a foundation for advancements in data assimilation and utilization.

Analysis of tunnel face stability with advanced pipes support

To keep the tunnel face stable is very important for tunnel construction.In this paper,the tunnel face stability under the advanced pipe was analyzed using the Winkler foundation model and rigid limit equilibrium.The tunnel face deformation characteristics were also analyzed using the numerical simulation.The influence of parameters on the deflection of the pipe roof and the stability of the tunnel face were discussed.The results show that the tunnel face stability can be improved through increasing the pipe diameter,decreasing the initial displacement at the beginning of the pipe seat,and adopting the short round length and small excavation height.With the increase of tunnel burial depth,the safety factor of tunnel face stability first decreases,then increases,and then remains unchanged.The deformation at the center of the tunnel face is larger than the deformation at the surround sides and at the corner.The horizontal displacement varies little with the increasing of the pipe length.The horizontal displacement at the center of the tunnel face increases with the increase of the pipe ring spacing and the pipe longitudinal spacing.There is an optimum external angle.

Influence of seismic input on response of Baihetan arch dam

The dynamic responses of the arch dam including dam-foundation-storage capacity of water system,using two different earthquake input models,i.e.viscous-spring artificial boundary(AB)condition and massless foundation(MF),were studied and analyzed for the 269 m high Baihetan arch dam under construction in China.By using different input models,the stress and opening of contraction joints(OCJs)of arch dam under strong shock were taken into consideration.The results show that the earthquake input models have slight influence on the responses including earthquake stresses and openings of contraction joints in different extents.

Gateside packwall design in solid backfill mining-A case study

Based upon characteristic movement features of the overlying strata in solid backfill mining and in-situ observations,an associated model representing a roadway support system has been developed.Based on the Winkler foundation and beam model,the current study presents a static analysis of the model,thus permitting acquisition of a theoretical formula pertaining to roof convergence.Through use of working face 6304-1(Jisan Colliery) as the research setting,the association between roof convergence magnitude and both packwall strength and width have been elucidated.Based upon observed conditions at the working face,realistic packwall parameters have been formulated,with numerical simulation results and field application results indicating that design parameters garnered from the developed formula successfully adapted to local geological movement and deformation.Accordingly,roadway deformation was shown to be within the permissible range,thus satisfying mine production requirements.The proposed method in the current study may give a design basis for pack design in the context of SBM under similar conditions.

Theoretical investigation of interaction between a rectangular plate and fractional viscoelastic foundation

The interaction between plates and foundations is a typical problem encountered in geotechnical engineering. The long-term plate performance is highly dependent on the theological characteristics of ground soil. Compared with conventional linear theology, the fractional calculus-based theory is a more powerful mathematical tool that can address this issue. This paper proposes a fractional Merchant model （FMM） to investigate the time-dependent behavior of a simply supported rectangular plate on viscoelastic foundation. The correspondence principle involving Laplace transforms was employed to derive the closed-form solutions of plate response under uniformly distributed load. The plate deflection, bending moment, and foundation reaction calculated using the FMM were compared with the results obtained from the analogous elastic model （EM） and the standard Merchant model （SMM）. It is shown that the upper and lower bound solutions of the FMM can be determined using the EM. In addition, a parametric study was performed to examine the influences of the model parameters on the time- dependent behavior of the plate-foundation interaction problem. The results indicate that a small fractional differential order corresponds to a plate resting on a sandy soil foundation, while the fractional differential order value should be increased for a clayey soil foundation. The long-term performance of a foundation plate can be accurately simulated by varying the values of the fractional differential order and the viscosity coefficient. The observations from this study reveal that the proposed fractional model has the capability to capture the variation of plate deflection over many decades of time.

Elastic Modulus Limit of Support Layer in Twin-Block Ballastless Track Based on Train Load Effect

The support layer is an important component of twin-block ballastless track. The modulus of the support layer is an important design parameter and must be carefully solved. We studied the bending stress and deformation of track slab and support layer due to train load using the beam-plate finite element model on elastic foundation. The results show that support layer type has great impact on both support layer deformation and the stress on subgrade, but has little impact on the bending stress of either track slab or support layer. The continuous support layer type, and articulated support layer type with shear transfer device at their ends, are recommended. In order to keep the stress in the support layer less than that in track slab, the modulus of the continuous, unit, and articulated types of support layer （ in unit twin-block ballastless track）, and the support layer in continuous twin-block ballastless track, should not be larger than 15, 22, 20.5 and 5 GPa, respectively. In addition, the modulus of the unit-type support layer should not be more than 20 GPa, to ensure the step in support layer remains less than 1 mm.

Enhanced elastic beam model with BADS integrated for settlement assessment of immersed tunnels

Excessive settlement may induce structural damage and water leakage in immersed tunnels,seriously threatening the tunnels’safety.However,making accurate assessment of the settlement in immersed tunnels is difficult due to the incomplete knowledge of the geotechnical parameters and the inadequacy of the model itself.This paper proposes an effective method to accurately assess the settlement in immersed tunnels.An enhanced beam on elastic foundation model(E-BEFM)is developed for the settlement assessment,with the Bayesian adaptive direct search algorithm adopted to estimate unknown model parameters based on previous observations.The proposed method is applied to a field case of the Hong Kong–Zhuhai–Macao immersed tunnel.The original BEFM is used for comparison to highlight the better assessment performance of E-BEFM,particularly for joints’differential settlement.Results show that the proposed method can provide accurate predictions of the total settlement,angular distortion(a representation of tubes’relatively differential settlement),and joints’differential settlement,which consequently supports the associated maintenance decision-making and potential risk prevention for immersed tunnels in service.

On Realization of Intelligent Decision Making in the Real World:A Foundation Decision Model Perspective

The pervasive uncertainty and dynamic nature of real-world environments present significant challenges for the widespread implementation of machine-driven Intelligent Decision-Making(IDM)systems.Consequently,IDM should possess the ability to continuously acquire new skills and effectively generalize across a broad range of applications.The advancement of Artificial General Intelligence(AGI)that transcends task and application boundaries is critical for enhancing IDM.Recent studies have extensively investigated the Transformer neural architecture as a foundational model for various tasks,including computer vision,natural language processing,and reinforcement learning.We propose that a Foundation Decision Model(FDM)can be developed by formulating diverse decision-making tasks as sequence decoding tasks using the Transformer architecture,offering a promising solution for expanding IDM applications in complex real-world situations.In this paper,we discuss the efficiency and generalization improvements offered by a foundation decision model for IDM and explore its potential applications in multi-agent game AI,production scheduling,and robotics tasks.Lastly,we present a case study demonstrating our FDM implementation,DigitalBrain(DB1)with 1.3 billion parameters,achieving human-level performance in 870 tasks,such as text generation,image captioning,video game playing,robotic control,and traveling salesman problems.As a foundation decision model,DB1 represents an initial step toward more autonomous and efficient real-world IDM applications.

Benefits, Difficulties, and Success Factors of Three Different Quality Assessment Processes in Estonia

Quality assessment systems for business organisations and also for vocational schools were established in Estonia at the beginning of 2000s. Almost ten years later, corresponding systems were introduced also for higher education institutions （HEIs）. All these assessment systems are based on the European Foundation of Quality Management （EFQM） excellence model. The aim of this paper is to analyse benefits, difficulties, and success factors of quality assessment processes in Estonian business organisations, vocational schools, and HEIs. The study is based on the analysis of feedback questionnaire of 404 representatives from participating organisations （HEIs, vocational schools, and business enterprises） and assessors. Our analysis revealed that quality assessment processes including self-assessment reporting had a positive effect on organisation development and it has given the participants certain benefits and new challenges. As a consequence, knowledge about quality management as well as the self-assessment skills has been improved. The main difficulties were connected to limited time resources and with problems to recognize direct benefits of quality assessment. As substantial difficulties, limitations of analytical and report-writing skills were considered. As the biggest success factor, an involvement of managers on different levels decision making was recognized. The analysis revealed that there are a number of similar benefits and difficulties in the quality assessment systems of different organizations.

Segment Anything Is Not Always Perfect: An Investigation of SAM on Different Real-world Applications

Recently,Meta AI Research approaches a general,promptable segment anything model(SAM)pre-trained on an unprecedentedly large segmentation dataset(SA-1B).Without a doubt,the emergence of SAM will yield significant benefits for a wide array of practical image segmentation applications.In this study,we conduct a series of intriguing investigations into the performance of SAM across various applications,particularly in the fields of natural images,agriculture,manufacturing,remote sensing and healthcare.We analyze and discuss the benefits and limitations of SAM,while also presenting an outlook on its future development in segmentation tasks.By doing so,we aim to give a comprehensive understanding of SAM's practical applications.This work is expected to provide insights that facilitate future research activities toward generic segmentation.Source code is publicly available at https://github.com/LiuTingWed/SAM-Not-Perfect.

An improved practical model for wear prediction of revolute clearance joints in crank slider mechanisms

A stationary clearance link algorithm(SCLA)for calculating the reaction-force of revolute clearance joints in crank slider mechanisms is proposed in this paper.The SCLA is more efficient than other algorithms of the same accuracy.Furthermore,based on the Winkler foundation model,an unsymmetrical Winkler foundation model and a double elastic layer Winkler model are proposed.By integrating a dynamic model and the unsymmetrical Winkler foundation model with Archard wear model,an improved integrated wear prediction model is also generated.A series of experiments have been performed to compare with the predicted analysis data,and the results showed a good agreement.As a real industry application,with the double elastic layer Winkler model,the integrated wear prediction model was successfully used to predict the wear depth of the joint bearing(bimetallic bearing)for the cantilever crane of a concrete pump truck of Sany Heavy Industry.