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.

BOUNDARY VALUE PROBLEMS OF CONJUGATE AND GENERALIZED K-HOLOMORPHIC FUNCTIONS IN C^(2)

In this paper,conjugate k-holomorphic functions and generalized k-holomorphic functions are defined in the two-dimensional complex space,and the corresponding Riemann boundary value problems and the inverse problems are discussed on generalized bicylinders.By the characteristics of the corresponding functions and boundary properties of the Cauchy type singular integral operators with conjugate k-holomorphic kernels,the general solutions and special solutions of the corresponding boundary value problems are studied in a detailed fashion,and the integral expressions of the solutions are obtained.

Identification of an immune-related gene signature for predicting prognosis and immunotherapy efficacy in liver cancer via cell-cell communication

BACKGROUND Liver cancer is one of the deadliest malignant tumors worldwide.Immunotherapy has provided hope to patients with advanced liver cancer,but only a small fraction of patients benefit from this treatment due to individual differences.Identifying immune-related gene signatures in liver cancer patients not only aids physicians in cancer diagnosis but also offers personalized treatment strategies,thereby improving patient survival rates.Although several methods have been developed to predict the prognosis and immunotherapeutic efficacy in patients with liver cancer,the impact of cell-cell interactions in the tumor microenvir-onment has not been adequately considered.AIM To identify immune-related gene signals for predicting liver cancer prognosis and immunotherapy efficacy.METHODS Cell grouping and cell-cell communication analysis were performed on single-cell RNA-sequencing data to identify highly active cell groups in immune-related pathways.Highly active immune cells were identified by intersecting the highly active cell groups with B cells and T cells.The significantly differentially expressed genes between highly active immune cells and other cells were subsequently selected as features,and a least absolute shrinkage and selection operator(LASSO)regression model was constructed to screen for diagnostic-related features.Fourteen genes that were selected more than 5 times in 10 LASSO regression experiments were included in a multivariable Cox regression model.Finally,3 genes(stathmin 1,cofilin 1,and C-C chemokine ligand 5)significantly associated with survival were identified and used to construct an immune-related gene signature.RESULTS The immune-related gene signature composed of stathmin 1,cofilin 1,and C-C chemokine ligand 5 was identified through cell-cell communication.The effectiveness of the identified gene signature was validated based on experi-mental results of predictive immunotherapy response,tumor mutation burden analysis,immune cell infiltration analysis,survival analysis,and expression analysis.CONCLUSION The findings suggest that the identified gene signature may contribute to a deeper understanding of the activity patterns of immune cells in the liver tumor microenvironment,providing insights for personalized treatment strategies.

THE GLOBAL EXISTENCE AND UNIQUENESS OF SMOOTH SOLUTIONS TO A FLUID-PARTICLE INTERACTION MODEL IN THE FLOWING REGIME

This paper is concerned with the Cauchy problem for a 3D fluid-particle interaction model in the so-called flowing regime inℝ3.Under the smallness assumption on both the external potential and the initial perturbation of the stationary solution in some Sobolev spaces,the existence and uniqueness of global smooth solutions in H3 of the system are established by using the careful energy method.

DYNAMIC ANALYSIS OF A TYPE OF FINANCIAL RISK CONTAGION MODEL INVOLVING IMMUNITY PERIOD AND SELF-RESCUE

In this paper,we study the dynamics of a Susceptible-Exposed-Infectious-Recovered(SEIR)nancial risk contagion model with time delay.Using stability theory and Hopf bifurcation theory,equilibria stability and Hopf bifurcation are analyzed in detail.Based on the epidemic model,we improve it by taking prior prevention and self-rescue into consideration,conclude pre-ventive intensity and self-rescue capabilities e ect the number of infections.At the same time,the analytical conditions for Hopf bifurcation are obtained,and the relevant results are veri ed by numerical simulations.

Necessary and Sufficient Conditions for Oscillations of the Generalized Liénard Systems

In this paper, we study the second-order nonlinear differential systems of Liénard-type x˙=1a(x)[ h(y)−F(x) ], y˙=−a(x)g(x). Necessary and sufficient conditions to ensure that all nontrivial solutions are oscillatory are established by using a new nonlinear integral inequality. Our results substantially extend and improve previous results known in the literature.

Teaching Reform of “Probability Theory and Mathematical Statistics” Under the Background of New Engineering

Probability theory and mathematical statistics are fundamental courses for various majors in science and engineering.In response to the current teaching situation,we should integrate theory with practice,implement teaching reform,and carry out teaching innovation.The article carries out blended teaching with deep integration of online and offline modes and within and outside of class,constructing innovative measures of“four integrations and four reshaping.”The article conducts diversified evaluations to stimulate learning motivation and help achieve talent cultivation goals.Through the close integration of probability theory and mathematical statistics course teaching with professional education and practical application,the“three-in-one”teaching goal of value shaping,ability cultivation,and knowledge exploration is achieved.The fundamental task of“cultivating morality and talents”is implemented.

High-Order Solitons and Hybrid Behavior of (3 + 1)-Dimensional Potential Yu-Toda-Sasa-Fukuyama Equation with Variable Coefficients

In this paper, some exact solutions of the (3 + 1)-dimensional variable-coefficient Yu-Toda-Sasa-Fukuyama equation are investigated. By using Hirota’s direct method and symbolic computation, we obtained N-soliton solution. By using the long wave limit method, the N-order rational solution can be obtained from N-order soliton solution. Then, through the paired complexification of parameters, the lump solution is obtained from N-order rational solution. Meanwhile, we obtained a hybrid solution between 1-lump solution and N-soliton (N=1,2) by using the long wave limit method and parameter complex. Furthermore, four different sets of three-dimensional graphs of solitons, lump solutions and hybrid solutions are drawn by selecting four different sets of coefficient functions which include one set of constant coefficient function and three sets of variable coefficient functions.

The Study of Root Subspace Decomposition between Characteristic Polynomials and Minimum Polynomial

Let Abe the linear transformation on the linear space V in the field P, Vλibe the root subspace corresponding to the characteristic polynomial of the eigenvalue λi, and Wλibe the root subspace corresponding to the minimum polynomial of λi. Consider the problem of whether Vλiand Wλiare equal under the condition that the characteristic polynomial of Ahas the same eigenvalue as the minimum polynomial (see Theorem 1, 2). This article uses the method of mutual inclusion to prove that Vλi=Wλi. Compared to previous studies and proofs, the results of this research can be directly cited in related works. For instance, they can be directly cited in Daoji Meng’s book “Introduction to Differential Geometry.”

INTERIOR ESTIMATES IN MORREY SPACES FOR SOLUTIONS OF ELLIPTIC EQUATIONS AND WEIGHTED BOUNDEDNESS FOR COMMUTATORS OF SINGULAR INTEGRAL OPERATORS

It is proved that, for the nondivergence elliptic equations Σi,jn=1aijuxixj=f, if f belongs to the generalized Morrey spaces Lp, (w), then uxixj ∈ Lp, (w), where u is the W2,p-solution of the equations. In order to obtain this, the author first establish the weighted boundedness for the commutators of some singular integral operators on Lp, (w).

THE CLASSIFICATION AND PERIOD OF STATES FOR MARKOV CHAIN IN RANDOM ENVIRONMENT

This paper is a continuation of [8]. In Section 1, three kinds of communication are introdnced for two states and the relations among them are investigated. In Section 2, two kinds of period of a state are introdnced and it is obtained that the period is a 'class property',i.e. two states x and y belong to same class implies the period of x is equal to the period of y.

Temperature-Dependent Debye Temperature and Specific Capacity of Graphene

The shot-range interaction and the atomic anharmonic vibration are both considered, and then the analytic functions of the Debye temperature, the specific capacity and the thermal conductivity of graphene with the temperature are obtained. The influence of anharmonic vibration on these thermal physical properties is also investigated. Some theoretical results are given. If only the harmonic approximation is considered, the Debye temperature of the graphene is unrelated to the temperature. If the anharmonic terms are considered, it increases slowly with the increasing temperature. The molar heat capacity of the graphene increases nonlinearly with the increasing temperature. The mean free path of phonons and the thermal conductivity of the graphene decrease nonlinearly with the increasing temperature. The relative changes of the Debye temperature, the specific heat capacity and the thermal conductivity caused by the anharmonic terms increase with the increasing temperature. The anharmonic effect of atomic vibration becomes more significant under higher temperature.

Verification and validation of detonation modeling

The mathematical model used to describe the detonation multi-physics phenomenon is usually given by highly coupled nonlinear partial differential equations. Numerical simulation and the computer aided engineering (CAE) technique has become the third pillar of detonation research, along with theory and experiment, due to the detonation phenomenon is difficult to explain by the theoretical analysis, and the cost required to accredit the reliability of detonation products is very high, even some physical experiments of detonation are impossible. The numerical simulation technique can solve these complex problems in the real situation repeatedly and reduce the design cost and time stunningly. But the reliability of numerical simulation software and the serviceability of the computational result seriously hinders the extension, application and the self-restoration of the simulation software, restricts its independently innovational ability. This article deals with the physical modeling, numerical simulation, and software development of detonation in a unified way. Verification and validation and uncertainty quantification (V&V&UQ) is an important approach in ensuring the credibility of the modeling and simulation of detonation. V&V of detonation is based on our independently developed detonation multiphysics software-LAD2D. We propose the verification method based on mathematical theory and program function as well as availability of its program execution. Validation is executed by comparing with the experiment data. At last, we propose the future prospect of numerical simulation software and the CAE technique, and we also pay attention to the research direction of V&V&UQ.

EXPECTED PRESENT VALUE OF TOTAL DIVIDENDS IN THE COMPOUND BINOMIAL MODEL WITH DELAYED CLAIMS AND RANDOM INCOME

In this paper, a compound binomial model with a constant dividend barrier and random income is considered. Two types of individual claims, main claims and by-claims, are defined, where every by-claim is induced by the main claim and may be delayed for one time period with a certain probability. The premium income is assumed to another binomial process to capture the uncertainty of the customer＇s arrivals and payments. A system of difference equations with certain boundary conditions for the expected present value of total dividend payments prior to ruin is derived and solved. Explicit results are obtained when the claim sizes are Kn distributed or the claim size distributions have finite support. Numerical results are also provided to illustrate the impact of the delay of by-claims on the expected present value of dividends.

A Remark on the Generalized Roper-Suffridge Extension Operator for Spirallike Mappings of Type β and Order α

Suppose f is a spirallike function of type β and order α on the unit disk D.Let Ωn,p1,p2,…,pn={z=（z2,z2,…,zn）′∈C^n：∑j=1^n｜zj）^Pj〈1},where 1≤p1≤2,pj≥1,j=2,…,n,are real numbers.In this paper,we will prove that Φn,β2,γ2,…βn,γn（f）（z）=（f（z1）, preserves spirallikeness of type β and order α on Ωn,p1,p2,…,Pn.

Classification and Reduction of Generalized Thin Film Equations

Classification and reduction of the generalized fourth-order nonlinear differential equations arising from theliquid films are considered.It is shown that these equations have solutions on subspaces of the polynomial,exponential ortrigonometric form defined by linear nth-order ordinary differential equations with constant coefficients for n=4,...,9.Several examples of exact solutions are presented.

GENERALIZATIONS AND CARTESIAN CLOSED SUBCATEGORIES OF SEMICONTINUOUS LATTICES

In this article, the authors mainly study how to obtain new semicontinuous lattices from the given semicontinuous lattices and discuss the conditions under which the image of a semicontinuous projection operator is also semicontinuous. Moreover, the authors investigate the relation between semicontinuous lattices and completely distributive lattices. Finally, it is proved that the strongly semicontinuous lattice category is a Cartesian closed category.

WEIGHTED BOUNDEDNESS OF PARAMETRIC MARCINKIEWICZ INTEGRAL AND HIGHER ORDER COMMUTATOR

In this paper, the weighted boundedness of parametric Marcinkiewicz integral and its commutator with rough kernels are considered. In addition, the weak type norm inequalities for parametric Marcinkiewicz integral and its commutator with different weight functions and Dini kernel are also discussed.

NEW SYSTEMS OF GENERALIZED QUASI-VARIATIONAL INCLUSIONS IN FC-SPACES AND APPLICATIONS

In this paper, we study some new systems of generalized quasi-variational inclusion problems in FC-spaces without convexity structure.By applying an existence theorem of maximal elements of set-valued mappings due to the author, some new existence theorems of solutions for the systems of generalized quasi-variational inclusion problems are proved in noncompact FC-spaces. As applications, some existence results of solutions for the system of quasi-optimization problems and mathematical programs with the systems of generalized quasi-variational inclusion constraints are obtained in FC-spaces.

Analyses of the Temporal Development and Yield Losses due to Sheath Blight of Rice (Rhizoctonia solani AG1.1a)

Sheath blight of rice, caused by Rhizoctonia solani AG 1. 1a, has become the most important disease and caused serious yield losses in some major rice-growing regions in China in recent years. In the present study, field plot experiment was conducted to examine the relationships between disease intensity and inoculum density （ID）, the seasonal disease epidemic dynamics, and yield reductions due to disease damages. Results from the experiment demonstrated that the areas under progress curves of disease severity and those of percent rice tillers diseased were positively and closely related to the relative initial ID of the pathogen. The inoculum density-disease （IDD） relationships were simulated and the impractical linear models were obtained. Both logistic and Gompertz functions could be used to simulate the disease progress dynamics in time, but the progress curves of the disease severity were modeled better by the Gompertz than by logistic function. However, the Richards function was found to be the best in simulating the disease progress curves when a most appropriate value was chosen for the shape parameter m by using the computer software Epitimulator. Sheath blight infection decreased rice yield very significantly and a yield reduction of 40% was recorded in rice crop with the highest inoculum density. Rice yield was linearly and negatively correlated with the disease severity and the percent tillers affected. The simulated models for all these relationships were computed through executing Epitimulator software and were presented in this paper.