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
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
Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstan...Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstanding performance across various domains,thereby prompting researchers to investigate their applicability in recommendation systems.However,due to the lack of task-specific knowledge and an inefficient feature extraction process,LLMs still have suboptimal performance in recommendation tasks.Therefore,external knowledge sources,such as knowledge graphs(KGs)and knowledge bases(KBs),are often introduced to address the issue of data sparsity.Compared to KGs,KBs possess higher retrieval efficiency,making them more suitable for scenarios where LLMs serve as recommenders.To this end,we introduce a novel framework integrating LLMs with KBs for enhanced retrieval generation,namely LLMKB.LLMKB initially leverages structured knowledge to create mapping dictionaries,extracting entity-relation information from heterogeneous knowledge to construct KBs.Then,LLMKB achieves the embedding calibration between user information representations and documents in KBs through retrieval model fine-tuning.Finally,LLMKB employs retrievalaugmented generation to produce recommendations based on fused text inputs,followed by post-processing.Experiment results on two public CRS datasets demonstrate the effectiveness of our framework.Our code is publicly available at the link:https://anonymous.4open.science/r/LLMKB-6FD0.展开更多
This study examines how foreign language education in the artificial intelligence(AI)era could assist the cultivation of national consciousness through a technology-enhanced pedagogy of film appreciation.Using The Wil...This study examines how foreign language education in the artificial intelligence(AI)era could assist the cultivation of national consciousness through a technology-enhanced pedagogy of film appreciation.Using The Wild Robot as a case study,we argue that cinematic narratives serve as cultural mirrors,offering immersive,reflective,and affective sites for intercultural learning.We propose a three-layered pedagogical framework-progressing from semiotic decoding,through narrative and value comparison,to creative identity construction-that integrates intelligent tools to develop both communicative competence and an agentive sense of belonging.The approach exemplifies a humanistic turn in language teaching,aiming to form“rooted global communicators”who can engage in cross-civilization dialogue with cultural confidence and critical awareness.展开更多
Anxiety,motivation,and strategy have long been seen as critical in second language acquisition.This study presents a systematic review of the literature on these variables in terms of their relationship with one anoth...Anxiety,motivation,and strategy have long been seen as critical in second language acquisition.This study presents a systematic review of the literature on these variables in terms of their relationship with one another,their effects on learning outcomes,and how they are affected by technology-assisted tools in the teaching of Chinese as a second language.This study includes 24 articles for the review study based on the criteria and process of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocol(PRISMA-P)and the clustering techniques of VOSviewer.It is found that 1)anxiety,motivation,and strategy were interrelated,that is,motivation was negatively associated with anxiety but positively related to strategy,while strategy could positively predict anxiety;2)anxiety could both positively and negatively affect learning outcomes,while motivation and strategy could both positively and insignificantly influence learning outcomes;3)the technology-assisted tools used in the classroom could both positively and negatively affect the levels of these variables and learning outcomes in the L2 Chinese context.The need to explore more complicated relationships between language-specific individual variables themselves and other possible factors that affect these variables,such as cultural ones,are also discussed for future research.展开更多
Mirror neuron system （MNS） represents one past decade, and it has been found to involve in multiple of the most important discoveries of cognitive neuroscience in the aspects of brain functions including action unde...Mirror neuron system （MNS） represents one past decade, and it has been found to involve in multiple of the most important discoveries of cognitive neuroscience in the aspects of brain functions including action understanding, imitation, language understanding, empathy, action prediction and speech evolution. This manuscript reviewed the function of MNS in action understanding as well as language evolution, and specifically assessed its roles as the bridge from body language to fluent speeches. Then we discussed the speech defects of autism patients due to the disruption of MNS. Finally, given that MNS is plastic in adult brain, we proposed MNS targeted therapy provides an efficient rehabilitation approach for brain damages conditions as well as autism patients.展开更多
ChatGPT is a powerful artificial intelligence(AI)language model that has demonstrated significant improvements in various natural language processing(NLP) tasks. However, like any technology, it presents potential sec...ChatGPT is a powerful artificial intelligence(AI)language model that has demonstrated significant improvements in various natural language processing(NLP) tasks. However, like any technology, it presents potential security risks that need to be carefully evaluated and addressed. In this survey, we provide an overview of the current state of research on security of using ChatGPT, with aspects of bias, disinformation, ethics, misuse,attacks and privacy. We review and discuss the literature on these topics and highlight open research questions and future directions.Through this survey, we aim to contribute to the academic discourse on AI security, enriching the understanding of potential risks and mitigations. We anticipate that this survey will be valuable for various stakeholders involved in AI development and usage, including AI researchers, developers, policy makers, and end-users.展开更多
Peptide-based therapeutics hold great promise for the treatment of various diseases;however,their clinical application is often hindered by toxicity challenges.The accurate prediction of peptide toxicity is crucial fo...Peptide-based therapeutics hold great promise for the treatment of various diseases;however,their clinical application is often hindered by toxicity challenges.The accurate prediction of peptide toxicity is crucial for designing safe peptide-based therapeutics.While traditional experimental approaches are time-consuming and expensive,computational methods have emerged as viable alternatives,including similarity-based and machine learning(ML)-/deep learning(DL)-based methods.However,existing methods often struggle with robustness and generalizability.To address these challenges,we propose HyPepTox-Fuse,a novel framework that fuses protein language model(PLM)-based embeddings with conventional descriptors.HyPepTox-Fuse integrates ensemble PLM-based embeddings to achieve richer peptide representations by leveraging a cross-modal multi-head attention mechanism and Transformer architecture.A robust feature ranking and selection pipeline further refines conventional descriptors,thus enhancing prediction performance.Our framework outperforms state-of-the-art methods in cross-validation and independent evaluations,offering a scalable and reliable tool for peptide toxicity prediction.Moreover,we conducted a case study to validate the robustness and generalizability of HyPepTox-Fuse,highlighting its effectiveness in enhancing model performance.Furthermore,the HyPepTox-Fuse server is freely accessible at https://balalab-skku.org/HyPepTox-Fuse/and the source code is publicly available at https://github.com/cbbl-skku-org/HyPepTox-Fuse/.The study thus presents an intuitive platform for predicting peptide toxicity and supports reproducibility through openly available datasets.展开更多
Recently,diffusion models have emerged as a promising paradigm for molecular design and optimization.However,most diffusion-based molecular generative models focus on modeling 2D graphs or 3D geom-etries,with limited ...Recently,diffusion models have emerged as a promising paradigm for molecular design and optimization.However,most diffusion-based molecular generative models focus on modeling 2D graphs or 3D geom-etries,with limited research on molecular sequence diffusion models.The International Union of Pure and Applied Chemistry(IUPAC)names are more akin to chemical natural language than the simplified molecular input line entry system(SMILES)for organic compounds.In this work,we apply an IUPAC-guided conditional diffusion model to facilitate molecular editing from chemical natural language to chemical language(SMILES)and explore whether the pre-trained generative performance of diffusion models can be transferred to chemical natural language.We propose DiffIUPAC,a controllable molecular editing diffusion model that converts IUPAC names to SMILES strings.Evaluation results demonstrate that our model out-performs existing methods and successfully captures the semantic rules of both chemical languages.Chemical space and scaffold analysis show that the model can generate similar compounds with diverse scaffolds within the specified constraints.Additionally,to illustrate the model’s applicability in drug design,we conducted case studies in functional group editing,analogue design and linker design.展开更多
In recent years,large language models(LLMs)have demonstrated immense potential in practical applications to enhance work efficiency and decision-making capabilities.However,specialized LLMs in the oil and gas engineer...In recent years,large language models(LLMs)have demonstrated immense potential in practical applications to enhance work efficiency and decision-making capabilities.However,specialized LLMs in the oil and gas engineering area are rarely developed.To aid in exploring and developing deep and ultra-deep unconventional reservoirs,there is a call for a personalized LLM on oil-and gas-related rock mechanics,which may handle complex professional data and make intelligent predictions and decisions.To that end,herein,we overview general and industry-specific LLMs.Then,a systematic workflow is proposed for building this domain-specific LLM for oil and gas engineering,including data collection and processing,model construction and training,model validation,and implementation in the specific domain.Moreover,three application scenarios are investigated:knowledge extraction from textural resources,field operation with multidisciplinary integration,and intelligent decision assistance.Finally,several challenges in developing this domain-specific LLM are highlighted.Our key findings are that geological surveys,laboratory experiments,field tests,and numerical simulations form the four original sources of rock mechanics data.Those data must flow through collection,storage,processing,and governance before being fed into LLM training.This domain-specific LLM can be trained by fine-tuning a general open-source LLM with professional data and constraints such as rock mechanics datasets and principles.The LLM can then follow the commonly used training and validation processes before being implemented in the oil and gas field.However,there are three primary challenges in building this domain-specific LLM:data standardization,data security and access,and striking a compromise between physics and data when building the model structure.Some of these challenges are administrative rather than technical,and overcoming those requires close collaboration between the different interested parties and various professional practitioners.展开更多
BACKGROUND Inflammatory bowel disease(IBD)is a global health burden that affects millions of individuals worldwide,necessitating extensive patient education.Large language models(LLMs)hold promise for addressing patie...BACKGROUND Inflammatory bowel disease(IBD)is a global health burden that affects millions of individuals worldwide,necessitating extensive patient education.Large language models(LLMs)hold promise for addressing patient information needs.However,LLM use to deliver accurate and comprehensible IBD-related medical information has yet to be thoroughly investigated.AIM To assess the utility of three LLMs(ChatGPT-4.0,Claude-3-Opus,and Gemini-1.5-Pro)as a reference point for patients with IBD.METHODS In this comparative study,two gastroenterology experts generated 15 IBD-related questions that reflected common patient concerns.These questions were used to evaluate the performance of the three LLMs.The answers provided by each model were independently assessed by three IBD-related medical experts using a Likert scale focusing on accuracy,comprehensibility,and correlation.Simultaneously,three patients were invited to evaluate the comprehensibility of their answers.Finally,a readability assessment was performed.RESULTS Overall,each of the LLMs achieved satisfactory levels of accuracy,comprehensibility,and completeness when answering IBD-related questions,although their performance varies.All of the investigated models demonstrated strengths in providing basic disease information such as IBD definition as well as its common symptoms and diagnostic methods.Nevertheless,when dealing with more complex medical advice,such as medication side effects,dietary adjustments,and complication risks,the quality of answers was inconsistent between the LLMs.Notably,Claude-3-Opus generated answers with better readability than the other two models.CONCLUSION LLMs have the potential as educational tools for patients with IBD;however,there are discrepancies between the models.Further optimization and the development of specialized models are necessary to ensure the accuracy and safety of the information provided.展开更多
In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilizati...In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilization of this information. This study proposes a novel framework for intelligent Question-and-Answer (Q&A) systems based on Retrieval-Augmented Generation (RAG) to address these issues. The system efficiently acquires domain-specific knowledge by leveraging external databases, including Relational Databases (RDBs) and graph databases, without additional fine-tuning for Large Language Models (LLMs). Crucially, the framework integrates a Dynamic Knowledge Base Updating Mechanism (DKBUM) and a Weighted Context-Aware Similarity (WCAS) method to enhance retrieval accuracy and mitigate inherent limitations of LLMs, such as hallucinations and lack of specialization. Additionally, the proposed DKBUM dynamically adjusts knowledge weights within the database, ensuring that the most recent and relevant information is utilized, while WCAS refines the alignment between queries and knowledge items by enhanced context understanding. Experimental validation demonstrates that the system can generate timely, accurate, and context-sensitive responses, making it a robust solution for managing complex business logic in specialized industries.展开更多
AIM:To assess the possibility of using different large language models(LLMs)in ocular surface diseases by selecting five different LLMS to test their accuracy in answering specialized questions related to ocular surfa...AIM:To assess the possibility of using different large language models(LLMs)in ocular surface diseases by selecting five different LLMS to test their accuracy in answering specialized questions related to ocular surface diseases:ChatGPT-4,ChatGPT-3.5,Claude 2,PaLM2,and SenseNova.METHODS:A group of experienced ophthalmology professors were asked to develop a 100-question singlechoice question on ocular surface diseases designed to assess the performance of LLMs and human participants in answering ophthalmology specialty exam questions.The exam includes questions on the following topics:keratitis disease(20 questions),keratoconus,keratomalaciac,corneal dystrophy,corneal degeneration,erosive corneal ulcers,and corneal lesions associated with systemic diseases(20 questions),conjunctivitis disease(20 questions),trachoma,pterygoid and conjunctival tumor diseases(20 questions),and dry eye disease(20 questions).Then the total score of each LLMs and compared their mean score,mean correlation,variance,and confidence were calculated.RESULTS:GPT-4 exhibited the highest performance in terms of LLMs.Comparing the average scores of the LLMs group with the four human groups,chief physician,attending physician,regular trainee,and graduate student,it was found that except for ChatGPT-4,the total score of the rest of the LLMs is lower than that of the graduate student group,which had the lowest score in the human group.Both ChatGPT-4 and PaLM2 were more likely to give exact and correct answers,giving very little chance of an incorrect answer.ChatGPT-4 showed higher credibility when answering questions,with a success rate of 59%,but gave the wrong answer to the question 28% of the time.CONCLUSION:GPT-4 model exhibits excellent performance in both answer relevance and confidence.PaLM2 shows a positive correlation(up to 0.8)in terms of answer accuracy during the exam.In terms of answer confidence,PaLM2 is second only to GPT4 and surpasses Claude 2,SenseNova,and GPT-3.5.Despite the fact that ocular surface disease is a highly specialized discipline,GPT-4 still exhibits superior performance,suggesting that its potential and ability to be applied in this field is enormous,perhaps with the potential to be a valuable resource for medical students and clinicians in the future.展开更多
In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operati...In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operations.However,several challenges have emerged within the high-speed railway dispatching and command system,including the heavy workload faced by dispatchers,the difficulty of quantifying subjective expertise,and the need for effective training of professionals.Amid the growing application of artificial intelligence technologies in railway systems,this study leverages Large Language Model(LLM)technology.LLMs bring enhanced intelligence,predictive capabilities,robust memory,and adaptability to diverse real-world scenarios.This study proposes a human-computer interactive intelligent scheduling auxiliary training system built on LLM technology.The system offers capabilities including natural dialogue,knowledge reasoning,and human feedback learning.With broad applicability,the system is suitable for vocational education,guided inquiry,knowledge-based Q&A,and other training scenarios.Validation results demonstrate its effectiveness in auxiliary training,providing substantial support for educators,students,and dispatching personnel in colleges and professional settings.展开更多
With direct expression of individual application domain patterns and ideas,domain-specific modeling language（DSML） is more and more frequently used to build models instead of using a combination of one or more gener...With direct expression of individual application domain patterns and ideas,domain-specific modeling language（DSML） is more and more frequently used to build models instead of using a combination of one or more general constructs.Based on the profile mechanism of unified modeling language（UML） 2.2,a kind of DSML is presented to model simulation testing systems of avionic software（STSAS）.To define the syntax,semantics and notions of the DSML,the domain model of the STSAS from which we generalize the domain concepts and relationships among these concepts is given,and then,the domain model is mapped into a UML meta-model,named UML-STSAS profile.Assuming a flight control system（FCS） as system under test（SUT）,we design the relevant STSAS.The results indicate that extending UML to the simulation testing domain can effectively and precisely model STSAS.展开更多
The rapid advancement of Artificial Intelligence(AI)and Large Language Models(LLMs)has led to their increasing integration into various domains,from text generation and translation to question-answering.However,a crit...The rapid advancement of Artificial Intelligence(AI)and Large Language Models(LLMs)has led to their increasing integration into various domains,from text generation and translation to question-answering.However,a critical question remains:do these sophisticated models,much like humans,exhibit susceptibility to cognitive biases?Understanding the presence and nature of such biases in AI is paramount for assessing their reliability,enhancing their performance,and predicting their societal impact.This research specifically investigates the susceptibility of Google’s Gemini 1.5 Pro and DeepSeek,two prominent LLMs,to framing effects and confirmation bias.The study meticulously designed a series of experimental trials,systematically manipulating information proportions and presentation orders to evaluate these biases.In the framing effect experiment,a genetic testing decision-making scenario was constructed.The proportion of positive and negative information(e.g.,20%,50%,or 80%positive)and their presentation order were varied.The models’inclination towards undergoing genetic testing was recorded.For the confirmation bias experiment,two reports-one positive and one negative-about“RoboTaxi”autonomous vehicles were provided.The proportion of erroneous information within these reports(10%,30%,and 50%)and their presentation order were systematically altered,and the models’support for each report was assessed.The findings demonstrate that both Gemini 1.5 Pro and DeepSeek are susceptible to framing effects.In the genetic testing scenario,their decision-making was primarily influenced by the proportion of positive and negative information presented.When the proportion of positive information was higher,both models showed a greater inclination to recommend or proceed with genetic testing.Conversely,a higher proportion of negative information led to greater caution or a tendency not to recommend the testing.Importantly,the order in which this information was presented did not significantly influence their decisions in the framing effect scenarios.Regarding confirmation bias,the two models exhibited distinct behaviors.Gemini 1.5 Pro did not show an overall preference for either positive or negative reports.However,its judgments were significantly influenced by the order of information presentation,demonstrating a“recency effect,”meaning it tended to support the report presented later.The proportion of erroneous information within the reports had no significant impact on Gemini 1.5 Pro’s decisions.In contrast,DeepSeek exhibited an overall confirmation bias,showing a clear preference for positive reports.Similar to Gemini 1.5 Pro,DeepSeek’s decisions were also significantly affected by the order of information presentation,while the proportion of misinformation had no significant effect.These results reveal human-like cognitive vulnerabilities in advanced LLMs,highlighting critical challenges to their reliability and objectivity in decision-making processes.Gemini 1.5 Pro’s sensitivity to presentation order and DeepSeek’s general preference for positive information,coupled with its sensitivity to order,underscore the need for careful evaluation of potential cognitive biases during the development and application of AI.The study suggests that effective measures are necessary to mitigate these biases and prevent potential negative societal impacts.Future research should include a broader range of models for comparative analysis and explore more complex interactive scenarios to further understand and address these phenomena.The findings contribute significantly to understanding the limitations and capabilities of current AI systems,guiding their responsible development,and anticipating their potential societal implications.展开更多
The automatic diagnosis of depression plays a crucial role in preventing the deterioration of depression symptoms.The interview-based method is the most wildly adopted technique in depression diagnosis.However,the siz...The automatic diagnosis of depression plays a crucial role in preventing the deterioration of depression symptoms.The interview-based method is the most wildly adopted technique in depression diagnosis.However,the size of the collected conversation data is limited,and the sample distributions from different participants usually differ drastically.These factors present a great challenge in building a decent deep learning model for automatic depression diagnosis.Recently,large language models have demonstrated impressive capabilities and achieved human-level performance in various tasks under zero-shot and few-shot scenarios.This sheds new light on the development of AI solutions for domainspecific tasks with limited data.In this paper,we propose a two-stage approach that exploits the current most capable and cost-effective language model,ChatGPT,to make a depression diagnosis on interview-based data.Specifically,in the first stage,we use ChatGPT to summarize the raw dialogue sample,thereby facilitating the extraction of depression-related information.In the second stage,we use ChatGPT to classify the summarised data to predict the depressed state of the sample.Our method can achieve approximately 76%accuracy with a text-only modality on the DAIC-WOZ dataset.In addition,our method outperforms the performance of the state-of-the-art model by 6.2%in the D4 dataset.Our work highlights the potential of using large language models for diagnosis-based depression diagnosis.展开更多
Sign language dataset is essential in sign language recognition and translation(SLRT). Current public sign language datasets are small and lack diversity, which does not meet the practical application requirements for...Sign language dataset is essential in sign language recognition and translation(SLRT). Current public sign language datasets are small and lack diversity, which does not meet the practical application requirements for SLRT. However, making a large-scale and diverse sign language dataset is difficult as sign language data on the Internet is scarce. In making a large-scale and diverse sign language dataset, some sign language data qualities are not up to standard. This paper proposes a two information streams transformer(TIST) model to judge whether the quality of sign language data is qualified. To verify that TIST effectively improves sign language recognition(SLR), we make two datasets, the screened dataset and the unscreened dataset. In this experiment, this paper uses visual alignment constraint(VAC) as the baseline model. The experimental results show that the screened dataset can achieve better word error rate(WER) than the unscreened dataset.展开更多
With the increasing of data on the internet, data analysis has become inescapable to gain time and efficiency, especially in bibliographic information retrieval systems. We can estimate the number of actual scientific...With the increasing of data on the internet, data analysis has become inescapable to gain time and efficiency, especially in bibliographic information retrieval systems. We can estimate the number of actual scientific journals points to around 40</span></span><span style="font-family:Verdana;"><span style="font-family:Verdana;"><span style="font-family:Verdana;">,</span></span></span><span><span><span style="font-family:""><span style="font-family:Verdana;">000 with about four million articles published each year. Machine learning and deep learning applied to recommender systems had become unavoidable whether in industry or in research. In this current, we propose an optimized interface for bibliographic information retrieval as a </span><span style="font-family:Verdana;">running example, which allows different kind of researchers to find their</span><span style="font-family:Verdana;"> needs following some relevant criteria through natural language understanding. Papers indexed in Web of Science and Scopus are in high demand. Natural language including text and linguistic-based techniques, such as tokenization, named entity recognition, syntactic and semantic analysis, are used to express natural language queries. Our Interface uses association rules to find more related papers for recommendation. Spanning trees are challenged to optimize the search process of the system.展开更多
We analyze the suitability of existing pre-trained transformer-based language models(PLMs)for abstractive text summarization on German technical healthcare texts.The study focuses on the multilingual capabilities of t...We analyze the suitability of existing pre-trained transformer-based language models(PLMs)for abstractive text summarization on German technical healthcare texts.The study focuses on the multilingual capabilities of these models and their ability to perform the task of abstractive text summarization in the healthcare field.The research hypothesis was that large language models could perform high-quality abstractive text summarization on German technical healthcare texts,even if the model is not specifically trained in that language.Through experiments,the research questions explore the performance of transformer language models in dealing with complex syntax constructs,the difference in performance between models trained in English and German,and the impact of translating the source text to English before conducting the summarization.We conducted an evaluation of four PLMs(GPT-3,a translation-based approach also utilizing GPT-3,a German language Model,and a domain-specific bio-medical model approach).The evaluation considered the informativeness using 3 types of metrics based on Recall-Oriented Understudy for Gisting Evaluation(ROUGE)and the quality of results which is manually evaluated considering 5 aspects.The results show that text summarization models could be used in the German healthcare domain and that domain-independent language models achieved the best results.The study proves that text summarization models can simplify the search for pre-existing German knowledge in various domains.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.
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Conversational recommender systems(CRSs)focus on refining preferences and providing personalized recommendations through natural language interactions and dialogue history.Large language models(LLMs)have shown outstanding performance across various domains,thereby prompting researchers to investigate their applicability in recommendation systems.However,due to the lack of task-specific knowledge and an inefficient feature extraction process,LLMs still have suboptimal performance in recommendation tasks.Therefore,external knowledge sources,such as knowledge graphs(KGs)and knowledge bases(KBs),are often introduced to address the issue of data sparsity.Compared to KGs,KBs possess higher retrieval efficiency,making them more suitable for scenarios where LLMs serve as recommenders.To this end,we introduce a novel framework integrating LLMs with KBs for enhanced retrieval generation,namely LLMKB.LLMKB initially leverages structured knowledge to create mapping dictionaries,extracting entity-relation information from heterogeneous knowledge to construct KBs.Then,LLMKB achieves the embedding calibration between user information representations and documents in KBs through retrieval model fine-tuning.Finally,LLMKB employs retrievalaugmented generation to produce recommendations based on fused text inputs,followed by post-processing.Experiment results on two public CRS datasets demonstrate the effectiveness of our framework.Our code is publicly available at the link:https://anonymous.4open.science/r/LLMKB-6FD0.
基金supported by the project:Hunan Provincial Educational Science Research Project“Research on Cultivating National Consciousness in College Foreign Language Courses(XJT23CGD001)”.
文摘This study examines how foreign language education in the artificial intelligence(AI)era could assist the cultivation of national consciousness through a technology-enhanced pedagogy of film appreciation.Using The Wild Robot as a case study,we argue that cinematic narratives serve as cultural mirrors,offering immersive,reflective,and affective sites for intercultural learning.We propose a three-layered pedagogical framework-progressing from semiotic decoding,through narrative and value comparison,to creative identity construction-that integrates intelligent tools to develop both communicative competence and an agentive sense of belonging.The approach exemplifies a humanistic turn in language teaching,aiming to form“rooted global communicators”who can engage in cross-civilization dialogue with cultural confidence and critical awareness.
文摘Anxiety,motivation,and strategy have long been seen as critical in second language acquisition.This study presents a systematic review of the literature on these variables in terms of their relationship with one another,their effects on learning outcomes,and how they are affected by technology-assisted tools in the teaching of Chinese as a second language.This study includes 24 articles for the review study based on the criteria and process of the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocol(PRISMA-P)and the clustering techniques of VOSviewer.It is found that 1)anxiety,motivation,and strategy were interrelated,that is,motivation was negatively associated with anxiety but positively related to strategy,while strategy could positively predict anxiety;2)anxiety could both positively and negatively affect learning outcomes,while motivation and strategy could both positively and insignificantly influence learning outcomes;3)the technology-assisted tools used in the classroom could both positively and negatively affect the levels of these variables and learning outcomes in the L2 Chinese context.The need to explore more complicated relationships between language-specific individual variables themselves and other possible factors that affect these variables,such as cultural ones,are also discussed for future research.
基金Sci-ence Foundation of Ministry of Education of China (No.FBB011469)
文摘Mirror neuron system （MNS） represents one past decade, and it has been found to involve in multiple of the most important discoveries of cognitive neuroscience in the aspects of brain functions including action understanding, imitation, language understanding, empathy, action prediction and speech evolution. This manuscript reviewed the function of MNS in action understanding as well as language evolution, and specifically assessed its roles as the bridge from body language to fluent speeches. Then we discussed the speech defects of autism patients due to the disruption of MNS. Finally, given that MNS is plastic in adult brain, we proposed MNS targeted therapy provides an efficient rehabilitation approach for brain damages conditions as well as autism patients.
文摘ChatGPT is a powerful artificial intelligence(AI)language model that has demonstrated significant improvements in various natural language processing(NLP) tasks. However, like any technology, it presents potential security risks that need to be carefully evaluated and addressed. In this survey, we provide an overview of the current state of research on security of using ChatGPT, with aspects of bias, disinformation, ethics, misuse,attacks and privacy. We review and discuss the literature on these topics and highlight open research questions and future directions.Through this survey, we aim to contribute to the academic discourse on AI security, enriching the understanding of potential risks and mitigations. We anticipate that this survey will be valuable for various stakeholders involved in AI development and usage, including AI researchers, developers, policy makers, and end-users.
基金supported by the National Research Foundation of Korea(NRF)funded by the Ministry of Science and ICT,Republic of Korea(Grant No.:RS-2024-00344752)supported by the Department of Integrative Biotechnology,Sungkyunkwan University(SKKU)and the BK21 FOUR Project,Republic of Korea.
文摘Peptide-based therapeutics hold great promise for the treatment of various diseases;however,their clinical application is often hindered by toxicity challenges.The accurate prediction of peptide toxicity is crucial for designing safe peptide-based therapeutics.While traditional experimental approaches are time-consuming and expensive,computational methods have emerged as viable alternatives,including similarity-based and machine learning(ML)-/deep learning(DL)-based methods.However,existing methods often struggle with robustness and generalizability.To address these challenges,we propose HyPepTox-Fuse,a novel framework that fuses protein language model(PLM)-based embeddings with conventional descriptors.HyPepTox-Fuse integrates ensemble PLM-based embeddings to achieve richer peptide representations by leveraging a cross-modal multi-head attention mechanism and Transformer architecture.A robust feature ranking and selection pipeline further refines conventional descriptors,thus enhancing prediction performance.Our framework outperforms state-of-the-art methods in cross-validation and independent evaluations,offering a scalable and reliable tool for peptide toxicity prediction.Moreover,we conducted a case study to validate the robustness and generalizability of HyPepTox-Fuse,highlighting its effectiveness in enhancing model performance.Furthermore,the HyPepTox-Fuse server is freely accessible at https://balalab-skku.org/HyPepTox-Fuse/and the source code is publicly available at https://github.com/cbbl-skku-org/HyPepTox-Fuse/.The study thus presents an intuitive platform for predicting peptide toxicity and supports reproducibility through openly available datasets.
基金supported by the Yonsei University graduate school Department of Integrative Biotechnology.
文摘Recently,diffusion models have emerged as a promising paradigm for molecular design and optimization.However,most diffusion-based molecular generative models focus on modeling 2D graphs or 3D geom-etries,with limited research on molecular sequence diffusion models.The International Union of Pure and Applied Chemistry(IUPAC)names are more akin to chemical natural language than the simplified molecular input line entry system(SMILES)for organic compounds.In this work,we apply an IUPAC-guided conditional diffusion model to facilitate molecular editing from chemical natural language to chemical language(SMILES)and explore whether the pre-trained generative performance of diffusion models can be transferred to chemical natural language.We propose DiffIUPAC,a controllable molecular editing diffusion model that converts IUPAC names to SMILES strings.Evaluation results demonstrate that our model out-performs existing methods and successfully captures the semantic rules of both chemical languages.Chemical space and scaffold analysis show that the model can generate similar compounds with diverse scaffolds within the specified constraints.Additionally,to illustrate the model’s applicability in drug design,we conducted case studies in functional group editing,analogue design and linker design.
基金supported by the National Natural Science Foundation of China(no.42277122)the Science Foun-dation of the China University of Petroleum,Beijing(No.2462024BJRC013).
文摘In recent years,large language models(LLMs)have demonstrated immense potential in practical applications to enhance work efficiency and decision-making capabilities.However,specialized LLMs in the oil and gas engineering area are rarely developed.To aid in exploring and developing deep and ultra-deep unconventional reservoirs,there is a call for a personalized LLM on oil-and gas-related rock mechanics,which may handle complex professional data and make intelligent predictions and decisions.To that end,herein,we overview general and industry-specific LLMs.Then,a systematic workflow is proposed for building this domain-specific LLM for oil and gas engineering,including data collection and processing,model construction and training,model validation,and implementation in the specific domain.Moreover,three application scenarios are investigated:knowledge extraction from textural resources,field operation with multidisciplinary integration,and intelligent decision assistance.Finally,several challenges in developing this domain-specific LLM are highlighted.Our key findings are that geological surveys,laboratory experiments,field tests,and numerical simulations form the four original sources of rock mechanics data.Those data must flow through collection,storage,processing,and governance before being fed into LLM training.This domain-specific LLM can be trained by fine-tuning a general open-source LLM with professional data and constraints such as rock mechanics datasets and principles.The LLM can then follow the commonly used training and validation processes before being implemented in the oil and gas field.However,there are three primary challenges in building this domain-specific LLM:data standardization,data security and access,and striking a compromise between physics and data when building the model structure.Some of these challenges are administrative rather than technical,and overcoming those requires close collaboration between the different interested parties and various professional practitioners.
基金Supported by the China Health Promotion Foundation Young Doctors'Research Foundation for Inflammatory Bowel Disease,the Taishan Scholars Program of Shandong Province,China,No.tsqn202306343National Natural Science Foundation of China,No.82270578.
文摘BACKGROUND Inflammatory bowel disease(IBD)is a global health burden that affects millions of individuals worldwide,necessitating extensive patient education.Large language models(LLMs)hold promise for addressing patient information needs.However,LLM use to deliver accurate and comprehensible IBD-related medical information has yet to be thoroughly investigated.AIM To assess the utility of three LLMs(ChatGPT-4.0,Claude-3-Opus,and Gemini-1.5-Pro)as a reference point for patients with IBD.METHODS In this comparative study,two gastroenterology experts generated 15 IBD-related questions that reflected common patient concerns.These questions were used to evaluate the performance of the three LLMs.The answers provided by each model were independently assessed by three IBD-related medical experts using a Likert scale focusing on accuracy,comprehensibility,and correlation.Simultaneously,three patients were invited to evaluate the comprehensibility of their answers.Finally,a readability assessment was performed.RESULTS Overall,each of the LLMs achieved satisfactory levels of accuracy,comprehensibility,and completeness when answering IBD-related questions,although their performance varies.All of the investigated models demonstrated strengths in providing basic disease information such as IBD definition as well as its common symptoms and diagnostic methods.Nevertheless,when dealing with more complex medical advice,such as medication side effects,dietary adjustments,and complication risks,the quality of answers was inconsistent between the LLMs.Notably,Claude-3-Opus generated answers with better readability than the other two models.CONCLUSION LLMs have the potential as educational tools for patients with IBD;however,there are discrepancies between the models.Further optimization and the development of specialized models are necessary to ensure the accuracy and safety of the information provided.
文摘In the context of power generation companies, vast amounts of specialized data and expert knowledge have been accumulated. However, challenges such as data silos and fragmented knowledge hinder the effective utilization of this information. This study proposes a novel framework for intelligent Question-and-Answer (Q&A) systems based on Retrieval-Augmented Generation (RAG) to address these issues. The system efficiently acquires domain-specific knowledge by leveraging external databases, including Relational Databases (RDBs) and graph databases, without additional fine-tuning for Large Language Models (LLMs). Crucially, the framework integrates a Dynamic Knowledge Base Updating Mechanism (DKBUM) and a Weighted Context-Aware Similarity (WCAS) method to enhance retrieval accuracy and mitigate inherent limitations of LLMs, such as hallucinations and lack of specialization. Additionally, the proposed DKBUM dynamically adjusts knowledge weights within the database, ensuring that the most recent and relevant information is utilized, while WCAS refines the alignment between queries and knowledge items by enhanced context understanding. Experimental validation demonstrates that the system can generate timely, accurate, and context-sensitive responses, making it a robust solution for managing complex business logic in specialized industries.
基金Supported by National Natural Science Foundation of China(No.82160195,No.82460203)Degree and Postgraduate Education Teaching Reform Project of Jiangxi Province(No.JXYJG-2020-026).
文摘AIM:To assess the possibility of using different large language models(LLMs)in ocular surface diseases by selecting five different LLMS to test their accuracy in answering specialized questions related to ocular surface diseases:ChatGPT-4,ChatGPT-3.5,Claude 2,PaLM2,and SenseNova.METHODS:A group of experienced ophthalmology professors were asked to develop a 100-question singlechoice question on ocular surface diseases designed to assess the performance of LLMs and human participants in answering ophthalmology specialty exam questions.The exam includes questions on the following topics:keratitis disease(20 questions),keratoconus,keratomalaciac,corneal dystrophy,corneal degeneration,erosive corneal ulcers,and corneal lesions associated with systemic diseases(20 questions),conjunctivitis disease(20 questions),trachoma,pterygoid and conjunctival tumor diseases(20 questions),and dry eye disease(20 questions).Then the total score of each LLMs and compared their mean score,mean correlation,variance,and confidence were calculated.RESULTS:GPT-4 exhibited the highest performance in terms of LLMs.Comparing the average scores of the LLMs group with the four human groups,chief physician,attending physician,regular trainee,and graduate student,it was found that except for ChatGPT-4,the total score of the rest of the LLMs is lower than that of the graduate student group,which had the lowest score in the human group.Both ChatGPT-4 and PaLM2 were more likely to give exact and correct answers,giving very little chance of an incorrect answer.ChatGPT-4 showed higher credibility when answering questions,with a success rate of 59%,but gave the wrong answer to the question 28% of the time.CONCLUSION:GPT-4 model exhibits excellent performance in both answer relevance and confidence.PaLM2 shows a positive correlation(up to 0.8)in terms of answer accuracy during the exam.In terms of answer confidence,PaLM2 is second only to GPT4 and surpasses Claude 2,SenseNova,and GPT-3.5.Despite the fact that ocular surface disease is a highly specialized discipline,GPT-4 still exhibits superior performance,suggesting that its potential and ability to be applied in this field is enormous,perhaps with the potential to be a valuable resource for medical students and clinicians in the future.
基金the Talent Fund of Beijing Jiaotong University(Grant No.2024XKRC055).
文摘In recent years,railway construction in China has developed vigorously.With continuous improvements in the highspeed railway network,the focus is gradually shifting from large-scale construction to large-scale operations.However,several challenges have emerged within the high-speed railway dispatching and command system,including the heavy workload faced by dispatchers,the difficulty of quantifying subjective expertise,and the need for effective training of professionals.Amid the growing application of artificial intelligence technologies in railway systems,this study leverages Large Language Model(LLM)technology.LLMs bring enhanced intelligence,predictive capabilities,robust memory,and adaptability to diverse real-world scenarios.This study proposes a human-computer interactive intelligent scheduling auxiliary training system built on LLM technology.The system offers capabilities including natural dialogue,knowledge reasoning,and human feedback learning.With broad applicability,the system is suitable for vocational education,guided inquiry,knowledge-based Q&A,and other training scenarios.Validation results demonstrate its effectiveness in auxiliary training,providing substantial support for educators,students,and dispatching personnel in colleges and professional settings.
基金Aeronautical Science Foundation of China (20095551025)
文摘With direct expression of individual application domain patterns and ideas,domain-specific modeling language（DSML） is more and more frequently used to build models instead of using a combination of one or more general constructs.Based on the profile mechanism of unified modeling language（UML） 2.2,a kind of DSML is presented to model simulation testing systems of avionic software（STSAS）.To define the syntax,semantics and notions of the DSML,the domain model of the STSAS from which we generalize the domain concepts and relationships among these concepts is given,and then,the domain model is mapped into a UML meta-model,named UML-STSAS profile.Assuming a flight control system（FCS） as system under test（SUT）,we design the relevant STSAS.The results indicate that extending UML to the simulation testing domain can effectively and precisely model STSAS.
文摘The rapid advancement of Artificial Intelligence(AI)and Large Language Models(LLMs)has led to their increasing integration into various domains,from text generation and translation to question-answering.However,a critical question remains:do these sophisticated models,much like humans,exhibit susceptibility to cognitive biases?Understanding the presence and nature of such biases in AI is paramount for assessing their reliability,enhancing their performance,and predicting their societal impact.This research specifically investigates the susceptibility of Google’s Gemini 1.5 Pro and DeepSeek,two prominent LLMs,to framing effects and confirmation bias.The study meticulously designed a series of experimental trials,systematically manipulating information proportions and presentation orders to evaluate these biases.In the framing effect experiment,a genetic testing decision-making scenario was constructed.The proportion of positive and negative information(e.g.,20%,50%,or 80%positive)and their presentation order were varied.The models’inclination towards undergoing genetic testing was recorded.For the confirmation bias experiment,two reports-one positive and one negative-about“RoboTaxi”autonomous vehicles were provided.The proportion of erroneous information within these reports(10%,30%,and 50%)and their presentation order were systematically altered,and the models’support for each report was assessed.The findings demonstrate that both Gemini 1.5 Pro and DeepSeek are susceptible to framing effects.In the genetic testing scenario,their decision-making was primarily influenced by the proportion of positive and negative information presented.When the proportion of positive information was higher,both models showed a greater inclination to recommend or proceed with genetic testing.Conversely,a higher proportion of negative information led to greater caution or a tendency not to recommend the testing.Importantly,the order in which this information was presented did not significantly influence their decisions in the framing effect scenarios.Regarding confirmation bias,the two models exhibited distinct behaviors.Gemini 1.5 Pro did not show an overall preference for either positive or negative reports.However,its judgments were significantly influenced by the order of information presentation,demonstrating a“recency effect,”meaning it tended to support the report presented later.The proportion of erroneous information within the reports had no significant impact on Gemini 1.5 Pro’s decisions.In contrast,DeepSeek exhibited an overall confirmation bias,showing a clear preference for positive reports.Similar to Gemini 1.5 Pro,DeepSeek’s decisions were also significantly affected by the order of information presentation,while the proportion of misinformation had no significant effect.These results reveal human-like cognitive vulnerabilities in advanced LLMs,highlighting critical challenges to their reliability and objectivity in decision-making processes.Gemini 1.5 Pro’s sensitivity to presentation order and DeepSeek’s general preference for positive information,coupled with its sensitivity to order,underscore the need for careful evaluation of potential cognitive biases during the development and application of AI.The study suggests that effective measures are necessary to mitigate these biases and prevent potential negative societal impacts.Future research should include a broader range of models for comparative analysis and explore more complex interactive scenarios to further understand and address these phenomena.The findings contribute significantly to understanding the limitations and capabilities of current AI systems,guiding their responsible development,and anticipating their potential societal implications.
基金supported by the Science and Technology Innovation 2030 Project of China(2021ZD0202600).
文摘The automatic diagnosis of depression plays a crucial role in preventing the deterioration of depression symptoms.The interview-based method is the most wildly adopted technique in depression diagnosis.However,the size of the collected conversation data is limited,and the sample distributions from different participants usually differ drastically.These factors present a great challenge in building a decent deep learning model for automatic depression diagnosis.Recently,large language models have demonstrated impressive capabilities and achieved human-level performance in various tasks under zero-shot and few-shot scenarios.This sheds new light on the development of AI solutions for domainspecific tasks with limited data.In this paper,we propose a two-stage approach that exploits the current most capable and cost-effective language model,ChatGPT,to make a depression diagnosis on interview-based data.Specifically,in the first stage,we use ChatGPT to summarize the raw dialogue sample,thereby facilitating the extraction of depression-related information.In the second stage,we use ChatGPT to classify the summarised data to predict the depressed state of the sample.Our method can achieve approximately 76%accuracy with a text-only modality on the DAIC-WOZ dataset.In addition,our method outperforms the performance of the state-of-the-art model by 6.2%in the D4 dataset.Our work highlights the potential of using large language models for diagnosis-based depression diagnosis.
基金supported by the National Language Commission to research on sign language data specifications for artificial intelligence applications and test standards for language service translation systems (No.ZDI145-70)。
文摘Sign language dataset is essential in sign language recognition and translation(SLRT). Current public sign language datasets are small and lack diversity, which does not meet the practical application requirements for SLRT. However, making a large-scale and diverse sign language dataset is difficult as sign language data on the Internet is scarce. In making a large-scale and diverse sign language dataset, some sign language data qualities are not up to standard. This paper proposes a two information streams transformer(TIST) model to judge whether the quality of sign language data is qualified. To verify that TIST effectively improves sign language recognition(SLR), we make two datasets, the screened dataset and the unscreened dataset. In this experiment, this paper uses visual alignment constraint(VAC) as the baseline model. The experimental results show that the screened dataset can achieve better word error rate(WER) than the unscreened dataset.
文摘With the increasing of data on the internet, data analysis has become inescapable to gain time and efficiency, especially in bibliographic information retrieval systems. We can estimate the number of actual scientific journals points to around 40</span></span><span style="font-family:Verdana;"><span style="font-family:Verdana;"><span style="font-family:Verdana;">,</span></span></span><span><span><span style="font-family:""><span style="font-family:Verdana;">000 with about four million articles published each year. Machine learning and deep learning applied to recommender systems had become unavoidable whether in industry or in research. In this current, we propose an optimized interface for bibliographic information retrieval as a </span><span style="font-family:Verdana;">running example, which allows different kind of researchers to find their</span><span style="font-family:Verdana;"> needs following some relevant criteria through natural language understanding. Papers indexed in Web of Science and Scopus are in high demand. Natural language including text and linguistic-based techniques, such as tokenization, named entity recognition, syntactic and semantic analysis, are used to express natural language queries. Our Interface uses association rules to find more related papers for recommendation. Spanning trees are challenged to optimize the search process of the system.
文摘We analyze the suitability of existing pre-trained transformer-based language models(PLMs)for abstractive text summarization on German technical healthcare texts.The study focuses on the multilingual capabilities of these models and their ability to perform the task of abstractive text summarization in the healthcare field.The research hypothesis was that large language models could perform high-quality abstractive text summarization on German technical healthcare texts,even if the model is not specifically trained in that language.Through experiments,the research questions explore the performance of transformer language models in dealing with complex syntax constructs,the difference in performance between models trained in English and German,and the impact of translating the source text to English before conducting the summarization.We conducted an evaluation of four PLMs(GPT-3,a translation-based approach also utilizing GPT-3,a German language Model,and a domain-specific bio-medical model approach).The evaluation considered the informativeness using 3 types of metrics based on Recall-Oriented Understudy for Gisting Evaluation(ROUGE)and the quality of results which is manually evaluated considering 5 aspects.The results show that text summarization models could be used in the German healthcare domain and that domain-independent language models achieved the best results.The study proves that text summarization models can simplify the search for pre-existing German knowledge in various domains.
