Recommendation systems are key to boosting user engagement,satisfaction,and retention,particularly on media platforms where personalized content is vital.Sequential recommendation systems learn from user-item interact...Recommendation systems are key to boosting user engagement,satisfaction,and retention,particularly on media platforms where personalized content is vital.Sequential recommendation systems learn from user-item interactions to predict future items of interest.However,many current methods rely on unique user and item IDs,limiting their ability to represent users and items effectively,especially in zero-shot learning scenarios where training data is scarce.With the rapid development of Large Language Models(LLMs),researchers are exploring their potential to enhance recommendation systems.However,there is a semantic gap between the linguistic semantics of LLMs and the collaborative semantics of recommendation systems,where items are typically indexed by IDs.Moreover,most research focuses on item representations,neglecting personalized user modeling.To address these issues,we propose a sequential recommendation framework using LLMs,called CIT-Rec,a model that integrates Collaborative semantics for user representation and Image and Text information for item representation to enhance Recommendations.Specifically,by aligning intuitive image information with text containing semantic features,we can more accurately represent items,improving item representation quality.We focus not only on item representations but also on user representations.To more precisely capture users’personalized preferences,we use traditional sequential recommendation models to train on users’historical interaction data,effectively capturing behavioral patterns.Finally,by combining LLMs and traditional sequential recommendation models,we allow the LLM to understand linguistic semantics while capturing collaborative semantics.Extensive evaluations on real-world datasets show that our model outperforms baseline methods,effectively combining user interaction history with item visual and textual modalities to provide personalized recommendations.展开更多
This study demonstrates a novel integration of large language models,machine learning,and multicriteria decision-making to investigate self-moderation in small online communities,a topic under-explored compared to use...This study demonstrates a novel integration of large language models,machine learning,and multicriteria decision-making to investigate self-moderation in small online communities,a topic under-explored compared to user behavior and platform-driven moderation on social media.The proposed methodological framework(1)utilizes large language models for social media post analysis and categorization,(2)employs k-means clustering for content characterization,and(3)incorporates the TODIM(Tomada de Decisão Interativa Multicritério)method to determine moderation strategies based on expert judgments.In general,the fully integrated framework leverages the strengths of these intelligent systems in a more systematic evaluation of large-scale decision problems.When applied in social media moderation,this approach promotes nuanced and context-sensitive self-moderation by taking into account factors such as cultural background and geographic location.The application of this framework is demonstrated within Facebook groups.Eight distinct content clusters encompassing safety,harassment,diversity,and misinformation are identified.Analysis revealed a preference for content removal across all clusters,suggesting a cautious approach towards potentially harmful content.However,the framework also highlights the use of other moderation actions,like account suspension,depending on the content category.These findings contribute to the growing body of research on self-moderation and offer valuable insights for creating safer and more inclusive online spaces within smaller communities.展开更多
Objectives:The five-year survival rate for pancreatic cancer is notably low,posing a significant challenge to patient health.The primary treatments are radiotherapy and chemotherapy,sometimes combined with targeted th...Objectives:The five-year survival rate for pancreatic cancer is notably low,posing a significant challenge to patient health.The primary treatments are radiotherapy and chemotherapy,sometimes combined with targeted therapy;however,their clinical benefits are limited.Therefore,developing new models to evaluate the therapeutic potential of novel molecules is essential.Fingolimod and Dimethyl Fumarate(DMF),currently used to treat multiple sclerosis,have recently been shown to have anti-cancer effects in several preclinical tumor models.This study aims to evaluate the therapeutic potential of Fingolimod and DMF in pancreatic cancer by investigating their respective in vitro cytotoxicity and in vivo antitumor effects.Methods:In this study,we evaluated for the first time these two drugs in pancreatic preclinical models in vitro using 3D spheroid tumor models and in vivo,which are compared to two standard-of-care consisting of Gemcitabine and Erlotinib.Results:In vitro,both Fingolimod and DMF induced cytotoxicity in spheroids from two pancreatic cell lines.Additionally,Fingolimod and DMF displayed anticancer effects in two subcutaneous xenograft models using PANC-1 and CFPAC-1 cells.Conclusions:Although the responses observed with Fingolimod and DMF were similar to those of Gemcitabine and Erlotinib,these findings indicate a potential emerging interest in Fingolimod and DMF for the treatment of pancreatic cancer.However,further work is still necessary to fully characterize how these drugs affect tumor progression.展开更多
Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Re...Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Recent approaches such as Distilling Step-by-Step(DSbS)introduce explanation supervision,yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation.In this work,we propose a task-structured curriculum learning(TSCL)framework that structures training into three sequential phases:(i)prediction-only,to establish stable feature representations;(ii)joint prediction-explanation,to align task outputs with rationale generation;and(iii)explanation-only,to refine the quality of rationales.This design provides a simple but effective modification to DSbS,requiring no architectural changes and adding negligible training cost.We justify the phase scheduling with ablation studies and convergence analysis,showing that an initial prediction-heavy stage followed by a balanced joint phase improves both stability and explanation alignment.Extensive experiments on five datasets(e-SNLI,ANLI,CommonsenseQA,SVAMP,and MedNLI)demonstrate that TSCL consistently outperforms strong baselines,achieving gains of+1.7-2.6 points in accuracy and 0.8-1.2 in ROUGE-L,corresponding to relative error reductions of up to 21%.Beyond lexical metrics,human evaluation and ERASERstyle faithfulness diagnostics confirm that TSCL produces more faithful and informative explanations.Comparative training curves further reveal faster convergence and lower variance across seeds.Efficiency analysis shows less than 3%overhead in wall-clock training time and no additional inference cost,making the approach practical for realworld deployment.This study demonstrates that a simple task-structured curriculum can significantly improve the effectiveness of knowledge distillation.By separating and sequencing objectives,TSCL achieves a better balance between accuracy,stability,and explanation quality.The framework generalizes across domains,including medical NLI,and offers a principled recipe for future applications in multimodal reasoning and reinforcement learning.展开更多
This study aims to develop an accurate and robust machine learning model to predict the carbonation depth of fly ash concrete,overcoming the limitations of traditional predictive methods.Five ensemble-based models,suc...This study aims to develop an accurate and robust machine learning model to predict the carbonation depth of fly ash concrete,overcoming the limitations of traditional predictive methods.Five ensemble-based models,such as adaptive boosting(AdaBoost),categorical boosting(CatBoost),gradient boosting regressor(GBR),hist gradient boosting regressor(HistGBR),and extreme gradient boosting(XGBoost),were developed and optimized using 729 high-quality dataset points incorporating seven input parameters,including cement,CO_(2),exposure time,water-binder ratio,fly ash,curing time,and compressive strength.Several performance evaluation metrics were used to compare the models.The GBR model emerged as the best-performing model,based on high coefficient of determination(R^(2))values and balanced error metrics across both validation and testing datasets.While all models performed exceptionally well on the training data,GBR demonstrated superior generalization capability,with R^(2) values of 0.9438 on the validation set and 0.9310 on the testing set.Furthermore,its low mean squared error(MSE),root mean square error(RMSE),mean absolute error(MAE),and median absolute error(MdAE)confirmed its robustness and accuracy.Moreover,shapley additive explanations(SHAP)analysis enhanced the interpretability of predictions,highlighting the curing time and exposure time as the most critical drivers of carbonation depth.展开更多
Purpose:ATLAS is a cross-sectional study aiming to investigate environmental and genetic determinants of athletic performance in healthy Greek competitive athletes(CA).This article presents the study design,investigat...Purpose:ATLAS is a cross-sectional study aiming to investigate environmental and genetic determinants of athletic performance in healthy Greek competitive athletes(CA).This article presents the study design,investigates the muscle strength performance(MSP)of 289 adult and teenage CA,exercisers,and physically inactive individuals(PI),and proposes predictive models of MSP for adults.Methods:Muscle maximal,speed,and explosive strength(MMS/MSS/MES)at unilateral maximal concentric flexion and extension contraction(FC/EC)were evaluated using Biodex System 3 PRO^(TM)at 60°/s,180°/s,and 300°/s,while additional performance markers were assessed through field ergometric testing.Participants were interviewed about their lifestyle,dietary habits,physical activity,injury,and medical history.Body composition was assessed via bioelectrical impedance.gDNA was extracted from biochemical samples and then genotyped.Statistical analysis was conducted using IBM SPSS Statistics v21.0 and R.Results:Age,fitness,and sex impacted correlations of MSP with body composition and anthropometric measurements(p<0.05).Among CA,females outperformed males in accuracy(p<0.001)while,males outperformed females in anaerobic power,MSP,speed,and endurance(p<0.001).Adult CA outperformed exercisers and PI in MMS,MSS,and MES(p<0.05).Multiple linear regression models,with predictors age,FFM,body extremity,training load explained the majority of variation in MMS(R^(2)_(adj):71.4%–88.9%),MSS(R^(2)_(adj):64.8%–78.4%),and MES(R^(2)_(adj):52.7%–68.4%)at EC,FC,and their mean(p<0.001).Conclusions:Muscle-strengthening strategies should be customized according to individual fitness levels,body composition,and anthropometric measurements.The innovative sex-specific regression models assessing MMS,MSS,and MES at EC and FC provide a framework for personalizing rehabilitation and skill-specific training strategies.展开更多
In this paper,we propose a new privacy-aware transmission scheduling algorithm for 6G ad hoc networks.This system enables end nodes to select the optimum time and scheme to transmit private data safely.In 6G dynamic h...In this paper,we propose a new privacy-aware transmission scheduling algorithm for 6G ad hoc networks.This system enables end nodes to select the optimum time and scheme to transmit private data safely.In 6G dynamic heterogeneous infrastructures,unstable links and non-uniform hardware capabilities create critical issues regarding security and privacy.Traditional protocols are often too computationally heavy to allow 6G services to achieve their expected Quality-of-Service(QoS).As the transport network is built of ad hoc nodes,there is no guarantee about their trustworthiness or behavior,and transversal functionalities are delegated to the extreme nodes.However,while security can be guaranteed in extreme-to-extreme solutions,privacy cannot,as all intermediate nodes still have to handle the data packets they are transporting.Besides,traditional schemes for private anonymous ad hoc communications are vulnerable against modern intelligent attacks based on learning models.The proposed scheme fulfills this gap.Findings show the probability of a successful intelligent attack reduces by up to 65%compared to ad hoc networks with no privacy protection strategy when used the proposed technology.While congestion probability can remain below 0.001%,as required in 6G services.展开更多
War rehearsals have become increasingly important in national security due to the growing complexity of international affairs.However,traditional rehearsal methods,such as military chess simulations,are inefficient an...War rehearsals have become increasingly important in national security due to the growing complexity of international affairs.However,traditional rehearsal methods,such as military chess simulations,are inefficient and inflexible,with particularly pronounced limitations in command and decision-making.The overwhelming volume of information and high decision complexity hinder the realization of autonomous and agile command and control.To address this challenge,an intelligent warfare simulation framework named Command-Agent is proposed,which deeply integrates large language models(LLMs)with digital twin battlefields.By constructing a highly realistic battlefield environment through real-time simulation and multi-source data fusion,the natural language interaction capabilities of LLMs are leveraged to lower the command threshold and to enable autonomous command through the Observe-Orient-Decide-Act(OODA)feedback loop.Within the Command-Agent framework,a multimodel collaborative architecture is further adopted to decouple the decision-generation and command-execution functions of LLMs.By combining specialized models such as Deep Seek-R1 and MCTool,the limitations of single-model capabilities are overcome.MCTool is a lightweight execution model fine-tuned for military Function Calling tasks.The framework also introduces a Vector Knowledge Base to mitigate hallucinations commonly exhibited by LLMs.Experimental results demonstrate that Command-Agent not only enables natural language-driven simulation and control but also deeply understands commander intent.Leveraging the multi-model collaborative architecture,during red-blue UAV confrontations involving 2 to 8 UAVs,the integrated score is improved by an average of 41.8%compared to the single-agent system(MCTool),accompanied by a 161.8%optimization in the battle loss ratio.Furthermore,when compared with multi-agent systems lacking the knowledge base,the inclusion of the Vector Knowledge Base further improves overall performance by 16.8%.In comparison with the general model(Qwen2.5-7B),the fine-tuned MCTool leads by 5%in execution efficiency.Therefore,the proposed Command-Agent introduces a novel perspective to the military command system and offers a feasible solution for intelligent battlefield decision-making.展开更多
Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that man...Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that manipulate model behavior through malicious instructions.Following Kitchenham’s guidelines,this systematic review synthesizes 128 peer-reviewed studies from 2022 to 2025 to provide a unified understanding of this rapidly evolving threat landscape.Our findings reveal a swift progression from simple direct injections to sophisticated multimodal attacks,achieving over 90%success rates against unprotected systems.In response,defense mechanisms show varying effectiveness:input preprocessing achieves 60%–80%detection rates and advanced architectural defenses demonstrate up to 95%protection against known patterns,though significant gaps persist against novel attack vectors.We identified 37 distinct defense approaches across three categories,but standardized evaluation frameworks remain limited.Our analysis attributes these vulnerabilities to fundamental LLM architectural limitations,such as the inability to distinguish instructions from data and attention mechanism vulnerabilities.This highlights critical research directions such as formal verification methods,standardized evaluation protocols,and architectural innovations for inherently secure LLM designs.展开更多
Pinus radiata(D.Don)dominates New Zealand's forestry industry,constituting 91%of plantations,and is among the world's most important plantation species.Given the socio-economic and environmental importance of ...Pinus radiata(D.Don)dominates New Zealand's forestry industry,constituting 91%of plantations,and is among the world's most important plantation species.Given the socio-economic and environmental importance of this species,it is important to have accurate and precise projections over time to make efficient decisions for forest management and greenfield investments in afforestation projects,especially for permanent carbon forests.Future projections of any natural resource systems rely on modeling;however,the acceleration of climate change makes future projections of yield less certain.These challenges also impact national expectations of the contribution planted forests will provide to address climate change and meet international commitments under the Paris Agreement.Using a large national-scale set of contemporary ground-measured data(2013–2023),this study investigates the performance of two growth models developed over 30 years ago that are widely used by NZ plantation growers:1)the Pumice Plateau Model 1988(PPM88)and 2)the 300-index(including a model variant of regional drift).Model simulations were made using the FORECASTER modeling suite with geographic boundaries to adjust for drift in space and time.Basal area(BA,m^(2)⋅ha^(-1))and volume(m^(3)⋅ha^(-1))were simulated,and standard errors and goodness-of-fit metrics calculated up to a typical rotation age of 30 years.Model residuals were then separated and analysed for the main plantation growing regions.The models overpredicted observed growth by between 6.8%and 16.2%,but model predictions and errors varied significantly between regions.The results of this study provided clear evidence of divergence between the outputs of both models and the measured data.Finally,this study suggests future measures to address challenges posed by these discrepancies that will provide better information for forest management and investment decisions in a changing climate.展开更多
Online Public Opinion Reports consolidate news and social media for timely crisis management by governments and enterprises.While large language models(LLMs)enable automated report generation,this specific domain lack...Online Public Opinion Reports consolidate news and social media for timely crisis management by governments and enterprises.While large language models(LLMs)enable automated report generation,this specific domain lacks formal task definitions and corresponding benchmarks.To bridge this gap,we define the Automated Online Public Opinion Report Generation(OPOR-Gen)task and construct OPOR-Bench,an event-centric dataset with 463 crisis events across 108 countries(comprising 8.8 K news articles and 185 K tweets).To evaluate report quality,we propose OPOR-Eval,a novel agent-based framework that simulates human expert evaluation.Validation experiments show OPOR-Eval achieves a high Spearman’s correlation(ρ=0.70)with human judgments,though challenges in temporal reasoning persist.This work establishes an initial foundation for advancing automated public opinion reporting research.展开更多
Myasthenia gravis is a chronic autoimmune disorder that affects the neuromuscular junction leading to fluctuating skeletal muscle fatigability. The majority of myasthenia gravis patients have detectable antibodies in ...Myasthenia gravis is a chronic autoimmune disorder that affects the neuromuscular junction leading to fluctuating skeletal muscle fatigability. The majority of myasthenia gravis patients have detectable antibodies in their serum, targeting acetylcholine receptor, muscle-specific kinase, or related proteins. Current treatment for myasthenia gravis involves symptomatic therapy, immunosuppressive drugs such as corticosteroids, azathioprine, and mycophenolate mofetil, and thymectomy, which is primarily indicated in patients with thymoma or thymic hyperplasia. However, this condition continues to pose significant challenges including an unpredictable and variable disease progression, differing response to individual therapies, and substantial longterm side effects associated with standard treatments(including an increased risk of infections, osteoporosis, and diabetes), underscoring the necessity for a more personalized approach to treatment. Furthermore, about fifteen percent of patients, called “refractory myasthenia gravis patients”, do not respond adequately to standard therapies. In this context, the introduction of molecular therapies has marked a significant advance in myasthenia gravis management. Advances in understanding myasthenia gravis pathogenesis, especially the role of pathogenic antibodies, have driven the development of these biological drugs, which offer more selective, rapid, and safer alternatives to traditional immunosuppressants. This review aims to provide a comprehensive overview of emerging therapeutic strategies targeting specific immune pathways in myasthenia gravis, with a particular focus on preclinical evidence, therapeutic rationale, and clinical translation of B-cell depletion therapies, neonatal Fc receptor inhibitors, and complement inhibitors.展开更多
The malicious dissemination of hate speech via compromised accounts,automated bot networks and malware-driven social media campaigns has become a growing cybersecurity concern.Automatically detecting such content in S...The malicious dissemination of hate speech via compromised accounts,automated bot networks and malware-driven social media campaigns has become a growing cybersecurity concern.Automatically detecting such content in Spanish is challenging due to linguistic complexity and the scarcity of annotated resources.In this paper,we compare two predominant AI-based approaches for the forensic detection of malicious hate speech:(1)finetuning encoder-only models that have been trained in Spanish and(2)In-Context Learning techniques(Zero-and Few-Shot Learning)with large-scale language models.Our approach goes beyond binary classification,proposing a comprehensive,multidimensional evaluation that labels each text by:(1)type of speech,(2)recipient,(3)level of intensity(ordinal)and(4)targeted group(multi-label).Performance is evaluated using an annotated Spanish corpus,standard metrics such as precision,recall and F1-score and stability-oriented metrics to evaluate the stability of the transition from zero-shot to few-shot prompting(Zero-to-Few Shot Retention and Zero-to-Few Shot Gain)are applied.The results indicate that fine-tuned encoder-only models(notably MarIA and BETO variants)consistently deliver the strongest and most reliable performance:in our experiments their macro F1-scores lie roughly in the range of approximately 46%–66%depending on the task.Zero-shot approaches are much less stable and typically yield substantially lower performance(observed F1-scores range approximately 0%–39%),often producing invalid outputs in practice.Few-shot prompting(e.g.,Qwen 38B,Mistral 7B)generally improves stability and recall relative to pure zero-shot,bringing F1-scores into a moderate range of approximately 20%–51%but still falling short of fully fine-tuned models.These findings highlight the importance of supervised adaptation and discuss the potential of both paradigms as components in AI-powered cybersecurity and malware forensics systems designed to identify and mitigate coordinated online hate campaigns.展开更多
Neurodegenerative disorders represent an increasingly pertinent public health crisis.As a greater proportion of the population ages,neurodegenerative disorders and other diseases of aging place undue burdens on patien...Neurodegenerative disorders represent an increasingly pertinent public health crisis.As a greater proportion of the population ages,neurodegenerative disorders and other diseases of aging place undue burdens on patients,caregivers,and healthcare workers.Alzheimer’s disease(AD)and Parkinson’s disease represent the two most common neurodegenerative disorders in the population,affecting over 65 million people,worldwide.展开更多
Background:Hepatocellular carcinoma(HCC)is a highly lethal malignancy driven by both intrinsic oncogenic pathways and immune microenvironmental regulation.Emerging evidence suggests that DNASE1L3 may influence tumor b...Background:Hepatocellular carcinoma(HCC)is a highly lethal malignancy driven by both intrinsic oncogenic pathways and immune microenvironmental regulation.Emerging evidence suggests that DNASE1L3 may influence tumor biology and immune responses;however,its specific roles in HCC progression and macrophage-mediated regulation remain unclear.This study aimed to elucidate the biological functions of DNASE1L3 in HCC and to determine how it modulates tumor behavior and immune interactions.Methods:Bioinformatics analyses of the GSE41804 and Cancer Genome Atlas-Liver Hepatocellular Carcinoma(TCGA-LIHC)datasets were used to identify hub genes.Functional assays assessed the impact of DNASE1L3 on HCC cell proliferation,migration,invasion,and cell cycle progression.The effects of DNASE1L3 on macrophage polarization and the Wnt/β-catenin signaling pathway were examined using a co-culture system.An HCC organoid model was established to further validate its regulatory function.Results:Eight prognostic signature genes were identified,with deoxyribonuclease I-like 3(DNase I-like 3)selected as the hub gene.DNASE1L3 overexpression suppressed HCC cell growth,inhibited migration and invasion,induced G1 arrest,and modulated epithelial-mesenchymal transition(EMT)markers.DNASE1L3 knockdown promoted M2-like macrophage polarization.Mechanistically,DNASE1L3 interacted withβ-catenin to enhance its ubiquitination and degradation,thereby inhibiting Wnt/β-catenin signaling and reducing PD-L1 expression.DNASE1L3 overexpression similarly restricted organoid growth and suppressed pathway activity.Conclusion:DNASE1L3 acts as a negative regulator of HCC progression by targeting the Wnt/β-catenin pathway and reducing PD-L1 expression,thereby influencing both tumor cell behavior and macrophage-mediated immune responses.展开更多
Gastrointestinal(GI)cancers represent a major global health concern due to their high incidence and mortality rates.Foundation models(FMs),also referred to as large models,represent a novel class of artificial intelli...Gastrointestinal(GI)cancers represent a major global health concern due to their high incidence and mortality rates.Foundation models(FMs),also referred to as large models,represent a novel class of artificial intelligence technologies that have demonstrated considerable potential in addressing these challenges.These models encompass large language models(LLMs),vision FMs(VFMs),and multimodal LLMs(MLLMs),all of which utilize transformer architectures and self-supervised pre-training on extensive unlabeled datasets to achieve robust cross-domain generalization.This review delineates the principal applications of these models:LLMs facilitate the structuring of clinical narratives,extraction of insights from medical records,and enhancement of physician-patient communication;VFMs are employed in the analysis of endoscopic,radiological,and pathological images for lesion detection and staging;MLLMs integrate heterogeneous data modalities,including imaging,textual information,and genomic data,to support diagnostic processes,treatment prediction,and prognostic evaluation.Despite these promising developments,several challenges remain,such as the need for data standardization,limited diversity within training datasets,substantial computational resource requirements,and ethical-legal concerns.In conclusion,FMs exhibit significant potential to advance research and clinical management of GI cancers.Future research efforts should prioritize the refinement of these models,promote international collaborations,and adopt interdisciplinary approaches.Such a comprehensive strategy is essential to fully harness the capabilities of FMs,driving substantial progress in the fight against GI malignancies.展开更多
In the era of AI,especially large models,the importance of open source has become increasingly prominent.First,open source allows innovation to avoid starting from scratch.Through iterative innovation,it promotes tech...In the era of AI,especially large models,the importance of open source has become increasingly prominent.First,open source allows innovation to avoid starting from scratch.Through iterative innovation,it promotes technical exchanges and learning globally.Second,resources required for large model R&D are difficult for a single institution to obtain.The evaluation of general large models also requires the participation of experts from various industries.Third,without open source collaboration,it is difficult to form a unified upper-layer software ecosystem.Therefore,open source has become an important cooperation mechanism to promote the development of AI and large models.There are two cases to illustrate how open source and international standards interact with each other.展开更多
Large language models(LLMs)have emerged as transformative tools in radiology artificial intelligence(AI),offering significant capabilities in areas such as image report generation,clinical decision support,and workflo...Large language models(LLMs)have emerged as transformative tools in radiology artificial intelligence(AI),offering significant capabilities in areas such as image report generation,clinical decision support,and workflow optimization.The first part of this manuscript presents a comprehensive overview of the current state of LLM applications in radiology,including their historical evolution,technical foundations,and practical uses.Despite notable advances,inherent architectural constraints,such as token-level sequential processing,limit their ability to perform deep abstract reasoning and holistic contextual understanding,which are critical for fine-grained diagnostic interpretation.We provide a critical perspective on current LLMs and discuss key challenges,including model reliability,bias,and explainability,highlighting the pressing need for novel approaches to advance radiology AI.Large concept models(LCMs)represent a nascent and promising paradigm in radiology AI,designed to transcend the limitations of token-level processing by utilizing higher-order conceptual representations and multimodal data integration.The second part of this manuscript introduces the foundational principles and theoretical framework of LCMs,highlighting their potential to facilitate enhanced semantic reasoning,long-range context synthesis,and improved clinical decision-making.Critically,the core of this section is the proposal of a novel theoretical framework for LCMs,formalized and extended from our group’s foundational concept-based models-the world’s earliest articulation of this paradigm for medical AI.This conceptual shift has since been externally validated and propelled by the recent publication of the LCM architectural proposal by Meta AI,providing a large-scale engineering blueprint for the future development of this technology.We also outline future research directions and the transformative implications of this emerging AI paradigm for radiologic practice,aiming to provide a blueprint for advancing toward human-like conceptual understanding in AI.While challenges persist,we are at the very beginning of a new era,and it is not unreasonable to hope that future advancements will overcome these hurdles,pushing the boundaries of AI in Radiology,far beyond even the most state-of-the-art models of today.展开更多
Lung cancer has one of the highest rates of incidence and mortality worldwide,mak-ing research on its mechanisms and treatments crucial.Animal models are essential in lung cancer research as they accurately replicate ...Lung cancer has one of the highest rates of incidence and mortality worldwide,mak-ing research on its mechanisms and treatments crucial.Animal models are essential in lung cancer research as they accurately replicate the biological characteristics and treatment outcomes seen in human diseases.Currently,various lung cancer models have been established,including chemical induction models,orthotopic transplan-tation models,ectopic transplantation models,metastasis models,and gene editing mouse models.Additionally,lung cancer grafts can be categorized into two types:tissue-based and cell-based grafts.This paper summarizes the phenotypes,advan-tages,and disadvantages of various induction methods based on their modeling tech-niques.The goal is to enhance the simulation of clinical lung cancer characteristics and to establish a solid foundation for future clinical research.展开更多
The application of visual-language large models in the field of medical health has gradually become a research focus.The models combine the capability for image understanding and natural language processing,and can si...The application of visual-language large models in the field of medical health has gradually become a research focus.The models combine the capability for image understanding and natural language processing,and can simultaneously process multi-modality data such as medical images and medical reports.These models can not only recognize images,but also understand the semantic relationship between images and texts,effectively realize the integration of medical information,and provide strong support for clinical decision-making and disease diagnosis.The visual-language large model has good performance for specific medical tasks,and also shows strong potential and high intelligence in the general task models.This paper provides a comprehensive review of the visual-language large model in the field of medical health.Specifically,this paper first introduces the basic theoretical basis and technical principles.Then,this paper introduces the specific application scenarios in the field of medical health,including modality fusion,semi-supervised learning,weakly supervised learning,unsupervised learning,cross-domain model and general models.Finally,the challenges including insufficient data,interpretability,and practical deployment are discussed.According to the existing challenges,four potential future development directions are given.展开更多
基金supported by the National Key R&D Program of China[2022YFF0902703]the State Administration for Market Regulation Science and Technology Plan Project(2024MK033).
文摘Recommendation systems are key to boosting user engagement,satisfaction,and retention,particularly on media platforms where personalized content is vital.Sequential recommendation systems learn from user-item interactions to predict future items of interest.However,many current methods rely on unique user and item IDs,limiting their ability to represent users and items effectively,especially in zero-shot learning scenarios where training data is scarce.With the rapid development of Large Language Models(LLMs),researchers are exploring their potential to enhance recommendation systems.However,there is a semantic gap between the linguistic semantics of LLMs and the collaborative semantics of recommendation systems,where items are typically indexed by IDs.Moreover,most research focuses on item representations,neglecting personalized user modeling.To address these issues,we propose a sequential recommendation framework using LLMs,called CIT-Rec,a model that integrates Collaborative semantics for user representation and Image and Text information for item representation to enhance Recommendations.Specifically,by aligning intuitive image information with text containing semantic features,we can more accurately represent items,improving item representation quality.We focus not only on item representations but also on user representations.To more precisely capture users’personalized preferences,we use traditional sequential recommendation models to train on users’historical interaction data,effectively capturing behavioral patterns.Finally,by combining LLMs and traditional sequential recommendation models,we allow the LLM to understand linguistic semantics while capturing collaborative semantics.Extensive evaluations on real-world datasets show that our model outperforms baseline methods,effectively combining user interaction history with item visual and textual modalities to provide personalized recommendations.
基金funded by the Office of the Vice-President for Research and Development of Cebu Technological University.
文摘This study demonstrates a novel integration of large language models,machine learning,and multicriteria decision-making to investigate self-moderation in small online communities,a topic under-explored compared to user behavior and platform-driven moderation on social media.The proposed methodological framework(1)utilizes large language models for social media post analysis and categorization,(2)employs k-means clustering for content characterization,and(3)incorporates the TODIM(Tomada de Decisão Interativa Multicritério)method to determine moderation strategies based on expert judgments.In general,the fully integrated framework leverages the strengths of these intelligent systems in a more systematic evaluation of large-scale decision problems.When applied in social media moderation,this approach promotes nuanced and context-sensitive self-moderation by taking into account factors such as cultural background and geographic location.The application of this framework is demonstrated within Facebook groups.Eight distinct content clusters encompassing safety,harassment,diversity,and misinformation are identified.Analysis revealed a preference for content removal across all clusters,suggesting a cautious approach towards potentially harmful content.However,the framework also highlights the use of other moderation actions,like account suspension,depending on the content category.These findings contribute to the growing body of research on self-moderation and offer valuable insights for creating safer and more inclusive online spaces within smaller communities.
基金supported by Porsolt SAS,https://www.porsolt.com/.
文摘Objectives:The five-year survival rate for pancreatic cancer is notably low,posing a significant challenge to patient health.The primary treatments are radiotherapy and chemotherapy,sometimes combined with targeted therapy;however,their clinical benefits are limited.Therefore,developing new models to evaluate the therapeutic potential of novel molecules is essential.Fingolimod and Dimethyl Fumarate(DMF),currently used to treat multiple sclerosis,have recently been shown to have anti-cancer effects in several preclinical tumor models.This study aims to evaluate the therapeutic potential of Fingolimod and DMF in pancreatic cancer by investigating their respective in vitro cytotoxicity and in vivo antitumor effects.Methods:In this study,we evaluated for the first time these two drugs in pancreatic preclinical models in vitro using 3D spheroid tumor models and in vivo,which are compared to two standard-of-care consisting of Gemcitabine and Erlotinib.Results:In vitro,both Fingolimod and DMF induced cytotoxicity in spheroids from two pancreatic cell lines.Additionally,Fingolimod and DMF displayed anticancer effects in two subcutaneous xenograft models using PANC-1 and CFPAC-1 cells.Conclusions:Although the responses observed with Fingolimod and DMF were similar to those of Gemcitabine and Erlotinib,these findings indicate a potential emerging interest in Fingolimod and DMF for the treatment of pancreatic cancer.However,further work is still necessary to fully characterize how these drugs affect tumor progression.
文摘Knowledge distillation has become a standard technique for compressing large language models into efficient student models,but existing methods often struggle to balance prediction accuracy with explanation quality.Recent approaches such as Distilling Step-by-Step(DSbS)introduce explanation supervision,yet they apply it in a uniform manner that may not fully exploit the different learning dynamics of prediction and explanation.In this work,we propose a task-structured curriculum learning(TSCL)framework that structures training into three sequential phases:(i)prediction-only,to establish stable feature representations;(ii)joint prediction-explanation,to align task outputs with rationale generation;and(iii)explanation-only,to refine the quality of rationales.This design provides a simple but effective modification to DSbS,requiring no architectural changes and adding negligible training cost.We justify the phase scheduling with ablation studies and convergence analysis,showing that an initial prediction-heavy stage followed by a balanced joint phase improves both stability and explanation alignment.Extensive experiments on five datasets(e-SNLI,ANLI,CommonsenseQA,SVAMP,and MedNLI)demonstrate that TSCL consistently outperforms strong baselines,achieving gains of+1.7-2.6 points in accuracy and 0.8-1.2 in ROUGE-L,corresponding to relative error reductions of up to 21%.Beyond lexical metrics,human evaluation and ERASERstyle faithfulness diagnostics confirm that TSCL produces more faithful and informative explanations.Comparative training curves further reveal faster convergence and lower variance across seeds.Efficiency analysis shows less than 3%overhead in wall-clock training time and no additional inference cost,making the approach practical for realworld deployment.This study demonstrates that a simple task-structured curriculum can significantly improve the effectiveness of knowledge distillation.By separating and sequencing objectives,TSCL achieves a better balance between accuracy,stability,and explanation quality.The framework generalizes across domains,including medical NLI,and offers a principled recipe for future applications in multimodal reasoning and reinforcement learning.
文摘This study aims to develop an accurate and robust machine learning model to predict the carbonation depth of fly ash concrete,overcoming the limitations of traditional predictive methods.Five ensemble-based models,such as adaptive boosting(AdaBoost),categorical boosting(CatBoost),gradient boosting regressor(GBR),hist gradient boosting regressor(HistGBR),and extreme gradient boosting(XGBoost),were developed and optimized using 729 high-quality dataset points incorporating seven input parameters,including cement,CO_(2),exposure time,water-binder ratio,fly ash,curing time,and compressive strength.Several performance evaluation metrics were used to compare the models.The GBR model emerged as the best-performing model,based on high coefficient of determination(R^(2))values and balanced error metrics across both validation and testing datasets.While all models performed exceptionally well on the training data,GBR demonstrated superior generalization capability,with R^(2) values of 0.9438 on the validation set and 0.9310 on the testing set.Furthermore,its low mean squared error(MSE),root mean square error(RMSE),mean absolute error(MAE),and median absolute error(MdAE)confirmed its robustness and accuracy.Moreover,shapley additive explanations(SHAP)analysis enhanced the interpretability of predictions,highlighting the curing time and exposure time as the most critical drivers of carbonation depth.
文摘Purpose:ATLAS is a cross-sectional study aiming to investigate environmental and genetic determinants of athletic performance in healthy Greek competitive athletes(CA).This article presents the study design,investigates the muscle strength performance(MSP)of 289 adult and teenage CA,exercisers,and physically inactive individuals(PI),and proposes predictive models of MSP for adults.Methods:Muscle maximal,speed,and explosive strength(MMS/MSS/MES)at unilateral maximal concentric flexion and extension contraction(FC/EC)were evaluated using Biodex System 3 PRO^(TM)at 60°/s,180°/s,and 300°/s,while additional performance markers were assessed through field ergometric testing.Participants were interviewed about their lifestyle,dietary habits,physical activity,injury,and medical history.Body composition was assessed via bioelectrical impedance.gDNA was extracted from biochemical samples and then genotyped.Statistical analysis was conducted using IBM SPSS Statistics v21.0 and R.Results:Age,fitness,and sex impacted correlations of MSP with body composition and anthropometric measurements(p<0.05).Among CA,females outperformed males in accuracy(p<0.001)while,males outperformed females in anaerobic power,MSP,speed,and endurance(p<0.001).Adult CA outperformed exercisers and PI in MMS,MSS,and MES(p<0.05).Multiple linear regression models,with predictors age,FFM,body extremity,training load explained the majority of variation in MMS(R^(2)_(adj):71.4%–88.9%),MSS(R^(2)_(adj):64.8%–78.4%),and MES(R^(2)_(adj):52.7%–68.4%)at EC,FC,and their mean(p<0.001).Conclusions:Muscle-strengthening strategies should be customized according to individual fitness levels,body composition,and anthropometric measurements.The innovative sex-specific regression models assessing MMS,MSS,and MES at EC and FC provide a framework for personalizing rehabilitation and skill-specific training strategies.
基金funding from the European Commission by the Ruralities project(grant agreement no.101060876).
文摘In this paper,we propose a new privacy-aware transmission scheduling algorithm for 6G ad hoc networks.This system enables end nodes to select the optimum time and scheme to transmit private data safely.In 6G dynamic heterogeneous infrastructures,unstable links and non-uniform hardware capabilities create critical issues regarding security and privacy.Traditional protocols are often too computationally heavy to allow 6G services to achieve their expected Quality-of-Service(QoS).As the transport network is built of ad hoc nodes,there is no guarantee about their trustworthiness or behavior,and transversal functionalities are delegated to the extreme nodes.However,while security can be guaranteed in extreme-to-extreme solutions,privacy cannot,as all intermediate nodes still have to handle the data packets they are transporting.Besides,traditional schemes for private anonymous ad hoc communications are vulnerable against modern intelligent attacks based on learning models.The proposed scheme fulfills this gap.Findings show the probability of a successful intelligent attack reduces by up to 65%compared to ad hoc networks with no privacy protection strategy when used the proposed technology.While congestion probability can remain below 0.001%,as required in 6G services.
文摘War rehearsals have become increasingly important in national security due to the growing complexity of international affairs.However,traditional rehearsal methods,such as military chess simulations,are inefficient and inflexible,with particularly pronounced limitations in command and decision-making.The overwhelming volume of information and high decision complexity hinder the realization of autonomous and agile command and control.To address this challenge,an intelligent warfare simulation framework named Command-Agent is proposed,which deeply integrates large language models(LLMs)with digital twin battlefields.By constructing a highly realistic battlefield environment through real-time simulation and multi-source data fusion,the natural language interaction capabilities of LLMs are leveraged to lower the command threshold and to enable autonomous command through the Observe-Orient-Decide-Act(OODA)feedback loop.Within the Command-Agent framework,a multimodel collaborative architecture is further adopted to decouple the decision-generation and command-execution functions of LLMs.By combining specialized models such as Deep Seek-R1 and MCTool,the limitations of single-model capabilities are overcome.MCTool is a lightweight execution model fine-tuned for military Function Calling tasks.The framework also introduces a Vector Knowledge Base to mitigate hallucinations commonly exhibited by LLMs.Experimental results demonstrate that Command-Agent not only enables natural language-driven simulation and control but also deeply understands commander intent.Leveraging the multi-model collaborative architecture,during red-blue UAV confrontations involving 2 to 8 UAVs,the integrated score is improved by an average of 41.8%compared to the single-agent system(MCTool),accompanied by a 161.8%optimization in the battle loss ratio.Furthermore,when compared with multi-agent systems lacking the knowledge base,the inclusion of the Vector Knowledge Base further improves overall performance by 16.8%.In comparison with the general model(Qwen2.5-7B),the fine-tuned MCTool leads by 5%in execution efficiency.Therefore,the proposed Command-Agent introduces a novel perspective to the military command system and offers a feasible solution for intelligent battlefield decision-making.
基金supported by 2023 Higher Education Scientific Research Planning Project of China Society of Higher Education(No.23PG0408)2023 Philosophy and Social Science Research Programs in Jiangsu Province(No.2023SJSZ0993)+2 种基金Nantong Science and Technology Project(No.JC2023070)Key Project of Jiangsu Province Education Science 14th Five-Year Plan(Grant No.B-b/2024/02/41)the Open Fund of Advanced Cryptography and System Security Key Laboratory of Sichuan Province(Grant No.SKLACSS-202407).
文摘Large language models(LLMs)have revolutionized AI applications across diverse domains.However,their widespread deployment has introduced critical security vulnerabilities,particularly prompt injection attacks that manipulate model behavior through malicious instructions.Following Kitchenham’s guidelines,this systematic review synthesizes 128 peer-reviewed studies from 2022 to 2025 to provide a unified understanding of this rapidly evolving threat landscape.Our findings reveal a swift progression from simple direct injections to sophisticated multimodal attacks,achieving over 90%success rates against unprotected systems.In response,defense mechanisms show varying effectiveness:input preprocessing achieves 60%–80%detection rates and advanced architectural defenses demonstrate up to 95%protection against known patterns,though significant gaps persist against novel attack vectors.We identified 37 distinct defense approaches across three categories,but standardized evaluation frameworks remain limited.Our analysis attributes these vulnerabilities to fundamental LLM architectural limitations,such as the inability to distinguish instructions from data and attention mechanism vulnerabilities.This highlights critical research directions such as formal verification methods,standardized evaluation protocols,and architectural innovations for inherently secure LLM designs.
基金funded by Scion's Strategic Science Investment Fund(SSIF)the Forest Growers Levy Trust(FGLT)through the Resilient Forests Programme(Task No.A89220)。
文摘Pinus radiata(D.Don)dominates New Zealand's forestry industry,constituting 91%of plantations,and is among the world's most important plantation species.Given the socio-economic and environmental importance of this species,it is important to have accurate and precise projections over time to make efficient decisions for forest management and greenfield investments in afforestation projects,especially for permanent carbon forests.Future projections of any natural resource systems rely on modeling;however,the acceleration of climate change makes future projections of yield less certain.These challenges also impact national expectations of the contribution planted forests will provide to address climate change and meet international commitments under the Paris Agreement.Using a large national-scale set of contemporary ground-measured data(2013–2023),this study investigates the performance of two growth models developed over 30 years ago that are widely used by NZ plantation growers:1)the Pumice Plateau Model 1988(PPM88)and 2)the 300-index(including a model variant of regional drift).Model simulations were made using the FORECASTER modeling suite with geographic boundaries to adjust for drift in space and time.Basal area(BA,m^(2)⋅ha^(-1))and volume(m^(3)⋅ha^(-1))were simulated,and standard errors and goodness-of-fit metrics calculated up to a typical rotation age of 30 years.Model residuals were then separated and analysed for the main plantation growing regions.The models overpredicted observed growth by between 6.8%and 16.2%,but model predictions and errors varied significantly between regions.The results of this study provided clear evidence of divergence between the outputs of both models and the measured data.Finally,this study suggests future measures to address challenges posed by these discrepancies that will provide better information for forest management and investment decisions in a changing climate.
基金supported by the Fundamental Research Funds for the Central Universities(No.CUC25SG013)the Foundation of Key Laboratory of Education Informatization for Nationalities(Yunnan Normal University),Ministry of Education(No.EIN2024C006).
文摘Online Public Opinion Reports consolidate news and social media for timely crisis management by governments and enterprises.While large language models(LLMs)enable automated report generation,this specific domain lacks formal task definitions and corresponding benchmarks.To bridge this gap,we define the Automated Online Public Opinion Report Generation(OPOR-Gen)task and construct OPOR-Bench,an event-centric dataset with 463 crisis events across 108 countries(comprising 8.8 K news articles and 185 K tweets).To evaluate report quality,we propose OPOR-Eval,a novel agent-based framework that simulates human expert evaluation.Validation experiments show OPOR-Eval achieves a high Spearman’s correlation(ρ=0.70)with human judgments,though challenges in temporal reasoning persist.This work establishes an initial foundation for advancing automated public opinion reporting research.
文摘Myasthenia gravis is a chronic autoimmune disorder that affects the neuromuscular junction leading to fluctuating skeletal muscle fatigability. The majority of myasthenia gravis patients have detectable antibodies in their serum, targeting acetylcholine receptor, muscle-specific kinase, or related proteins. Current treatment for myasthenia gravis involves symptomatic therapy, immunosuppressive drugs such as corticosteroids, azathioprine, and mycophenolate mofetil, and thymectomy, which is primarily indicated in patients with thymoma or thymic hyperplasia. However, this condition continues to pose significant challenges including an unpredictable and variable disease progression, differing response to individual therapies, and substantial longterm side effects associated with standard treatments(including an increased risk of infections, osteoporosis, and diabetes), underscoring the necessity for a more personalized approach to treatment. Furthermore, about fifteen percent of patients, called “refractory myasthenia gravis patients”, do not respond adequately to standard therapies. In this context, the introduction of molecular therapies has marked a significant advance in myasthenia gravis management. Advances in understanding myasthenia gravis pathogenesis, especially the role of pathogenic antibodies, have driven the development of these biological drugs, which offer more selective, rapid, and safer alternatives to traditional immunosuppressants. This review aims to provide a comprehensive overview of emerging therapeutic strategies targeting specific immune pathways in myasthenia gravis, with a particular focus on preclinical evidence, therapeutic rationale, and clinical translation of B-cell depletion therapies, neonatal Fc receptor inhibitors, and complement inhibitors.
基金the research project LaTe4PoliticES(PID2022-138099OB-I00)funded by MCIN/AEI/10.13039/501100011033 and the European Fund for Regional Development(ERDF)-a way to make Europe.Tomás Bernal-Beltrán is supported by University of Murcia through the predoctoral programme.
文摘The malicious dissemination of hate speech via compromised accounts,automated bot networks and malware-driven social media campaigns has become a growing cybersecurity concern.Automatically detecting such content in Spanish is challenging due to linguistic complexity and the scarcity of annotated resources.In this paper,we compare two predominant AI-based approaches for the forensic detection of malicious hate speech:(1)finetuning encoder-only models that have been trained in Spanish and(2)In-Context Learning techniques(Zero-and Few-Shot Learning)with large-scale language models.Our approach goes beyond binary classification,proposing a comprehensive,multidimensional evaluation that labels each text by:(1)type of speech,(2)recipient,(3)level of intensity(ordinal)and(4)targeted group(multi-label).Performance is evaluated using an annotated Spanish corpus,standard metrics such as precision,recall and F1-score and stability-oriented metrics to evaluate the stability of the transition from zero-shot to few-shot prompting(Zero-to-Few Shot Retention and Zero-to-Few Shot Gain)are applied.The results indicate that fine-tuned encoder-only models(notably MarIA and BETO variants)consistently deliver the strongest and most reliable performance:in our experiments their macro F1-scores lie roughly in the range of approximately 46%–66%depending on the task.Zero-shot approaches are much less stable and typically yield substantially lower performance(observed F1-scores range approximately 0%–39%),often producing invalid outputs in practice.Few-shot prompting(e.g.,Qwen 38B,Mistral 7B)generally improves stability and recall relative to pure zero-shot,bringing F1-scores into a moderate range of approximately 20%–51%but still falling short of fully fine-tuned models.These findings highlight the importance of supervised adaptation and discuss the potential of both paradigms as components in AI-powered cybersecurity and malware forensics systems designed to identify and mitigate coordinated online hate campaigns.
基金supported by the Canadian Institutes of Health Research(DFD-181599)the National Institutes of Health(T32AG058527)to RJB and R0190106435 to VM.
文摘Neurodegenerative disorders represent an increasingly pertinent public health crisis.As a greater proportion of the population ages,neurodegenerative disorders and other diseases of aging place undue burdens on patients,caregivers,and healthcare workers.Alzheimer’s disease(AD)and Parkinson’s disease represent the two most common neurodegenerative disorders in the population,affecting over 65 million people,worldwide.
基金funded by Shanghai Science and Technology Innovation Action Plan Project(22140901100)Shanghai Key Laboratory of Molecular Imaging(18DZ2260400)Shanghai University of Medicine and Health Science Seed Fund(SSF-24-21-01).
文摘Background:Hepatocellular carcinoma(HCC)is a highly lethal malignancy driven by both intrinsic oncogenic pathways and immune microenvironmental regulation.Emerging evidence suggests that DNASE1L3 may influence tumor biology and immune responses;however,its specific roles in HCC progression and macrophage-mediated regulation remain unclear.This study aimed to elucidate the biological functions of DNASE1L3 in HCC and to determine how it modulates tumor behavior and immune interactions.Methods:Bioinformatics analyses of the GSE41804 and Cancer Genome Atlas-Liver Hepatocellular Carcinoma(TCGA-LIHC)datasets were used to identify hub genes.Functional assays assessed the impact of DNASE1L3 on HCC cell proliferation,migration,invasion,and cell cycle progression.The effects of DNASE1L3 on macrophage polarization and the Wnt/β-catenin signaling pathway were examined using a co-culture system.An HCC organoid model was established to further validate its regulatory function.Results:Eight prognostic signature genes were identified,with deoxyribonuclease I-like 3(DNase I-like 3)selected as the hub gene.DNASE1L3 overexpression suppressed HCC cell growth,inhibited migration and invasion,induced G1 arrest,and modulated epithelial-mesenchymal transition(EMT)markers.DNASE1L3 knockdown promoted M2-like macrophage polarization.Mechanistically,DNASE1L3 interacted withβ-catenin to enhance its ubiquitination and degradation,thereby inhibiting Wnt/β-catenin signaling and reducing PD-L1 expression.DNASE1L3 overexpression similarly restricted organoid growth and suppressed pathway activity.Conclusion:DNASE1L3 acts as a negative regulator of HCC progression by targeting the Wnt/β-catenin pathway and reducing PD-L1 expression,thereby influencing both tumor cell behavior and macrophage-mediated immune responses.
基金Supported by the Open Project Program of Panxi Crops Research and Utilization Key Laboratory of Sichuan Province,No.SZKF202302the Fundamental Research Funds for the Central Universities No.2019CDYGYB024.
文摘Gastrointestinal(GI)cancers represent a major global health concern due to their high incidence and mortality rates.Foundation models(FMs),also referred to as large models,represent a novel class of artificial intelligence technologies that have demonstrated considerable potential in addressing these challenges.These models encompass large language models(LLMs),vision FMs(VFMs),and multimodal LLMs(MLLMs),all of which utilize transformer architectures and self-supervised pre-training on extensive unlabeled datasets to achieve robust cross-domain generalization.This review delineates the principal applications of these models:LLMs facilitate the structuring of clinical narratives,extraction of insights from medical records,and enhancement of physician-patient communication;VFMs are employed in the analysis of endoscopic,radiological,and pathological images for lesion detection and staging;MLLMs integrate heterogeneous data modalities,including imaging,textual information,and genomic data,to support diagnostic processes,treatment prediction,and prognostic evaluation.Despite these promising developments,several challenges remain,such as the need for data standardization,limited diversity within training datasets,substantial computational resource requirements,and ethical-legal concerns.In conclusion,FMs exhibit significant potential to advance research and clinical management of GI cancers.Future research efforts should prioritize the refinement of these models,promote international collaborations,and adopt interdisciplinary approaches.Such a comprehensive strategy is essential to fully harness the capabilities of FMs,driving substantial progress in the fight against GI malignancies.
文摘In the era of AI,especially large models,the importance of open source has become increasingly prominent.First,open source allows innovation to avoid starting from scratch.Through iterative innovation,it promotes technical exchanges and learning globally.Second,resources required for large model R&D are difficult for a single institution to obtain.The evaluation of general large models also requires the participation of experts from various industries.Third,without open source collaboration,it is difficult to form a unified upper-layer software ecosystem.Therefore,open source has become an important cooperation mechanism to promote the development of AI and large models.There are two cases to illustrate how open source and international standards interact with each other.
文摘Large language models(LLMs)have emerged as transformative tools in radiology artificial intelligence(AI),offering significant capabilities in areas such as image report generation,clinical decision support,and workflow optimization.The first part of this manuscript presents a comprehensive overview of the current state of LLM applications in radiology,including their historical evolution,technical foundations,and practical uses.Despite notable advances,inherent architectural constraints,such as token-level sequential processing,limit their ability to perform deep abstract reasoning and holistic contextual understanding,which are critical for fine-grained diagnostic interpretation.We provide a critical perspective on current LLMs and discuss key challenges,including model reliability,bias,and explainability,highlighting the pressing need for novel approaches to advance radiology AI.Large concept models(LCMs)represent a nascent and promising paradigm in radiology AI,designed to transcend the limitations of token-level processing by utilizing higher-order conceptual representations and multimodal data integration.The second part of this manuscript introduces the foundational principles and theoretical framework of LCMs,highlighting their potential to facilitate enhanced semantic reasoning,long-range context synthesis,and improved clinical decision-making.Critically,the core of this section is the proposal of a novel theoretical framework for LCMs,formalized and extended from our group’s foundational concept-based models-the world’s earliest articulation of this paradigm for medical AI.This conceptual shift has since been externally validated and propelled by the recent publication of the LCM architectural proposal by Meta AI,providing a large-scale engineering blueprint for the future development of this technology.We also outline future research directions and the transformative implications of this emerging AI paradigm for radiologic practice,aiming to provide a blueprint for advancing toward human-like conceptual understanding in AI.While challenges persist,we are at the very beginning of a new era,and it is not unreasonable to hope that future advancements will overcome these hurdles,pushing the boundaries of AI in Radiology,far beyond even the most state-of-the-art models of today.
基金Sichuan Provincial Administration of Traditional Chinese Medicine,Grant/Award Number:2023MS564National Natural Science Foundation of China,Grant/Award Number:82474436。
文摘Lung cancer has one of the highest rates of incidence and mortality worldwide,mak-ing research on its mechanisms and treatments crucial.Animal models are essential in lung cancer research as they accurately replicate the biological characteristics and treatment outcomes seen in human diseases.Currently,various lung cancer models have been established,including chemical induction models,orthotopic transplan-tation models,ectopic transplantation models,metastasis models,and gene editing mouse models.Additionally,lung cancer grafts can be categorized into two types:tissue-based and cell-based grafts.This paper summarizes the phenotypes,advan-tages,and disadvantages of various induction methods based on their modeling tech-niques.The goal is to enhance the simulation of clinical lung cancer characteristics and to establish a solid foundation for future clinical research.
基金The Natural Science Foundation of Hebei Province(F2024501044).
文摘The application of visual-language large models in the field of medical health has gradually become a research focus.The models combine the capability for image understanding and natural language processing,and can simultaneously process multi-modality data such as medical images and medical reports.These models can not only recognize images,but also understand the semantic relationship between images and texts,effectively realize the integration of medical information,and provide strong support for clinical decision-making and disease diagnosis.The visual-language large model has good performance for specific medical tasks,and also shows strong potential and high intelligence in the general task models.This paper provides a comprehensive review of the visual-language large model in the field of medical health.Specifically,this paper first introduces the basic theoretical basis and technical principles.Then,this paper introduces the specific application scenarios in the field of medical health,including modality fusion,semi-supervised learning,weakly supervised learning,unsupervised learning,cross-domain model and general models.Finally,the challenges including insufficient data,interpretability,and practical deployment are discussed.According to the existing challenges,four potential future development directions are given.
