10 April 2026, Volume 46 Issue 4
-
Select all|
-
LIBRARY TRIBUNE. 2026, 46(4): 1-11.By clarifying the connotations and characteristics of immersive public cultural services and defining the research object,this paper provides a systematic review of relevant domestic and international studies on the elements that constitute such services. Six core elements are identified：immersive experience subjects,immersive devices,immersive technologies,immersive experience spaces,immersive experience content,and immersive experience scenarios. The study also summarizes their sub-dimensions and characteristics. Focusing on the roles of these elements within the service system and their interrelationships,the paper establishes a hierarchical service framework comprising the support and guarantee layer,the experience development layer,and the service implementation layer. After exploring the application of this framework in practical settings,the study proposes recommendations based on influence mechanisms across different levels of immersion.
-
LIBRARY TRIBUNE. 2026, 46(4): 12-25.In the era of digital intelligence,human-intelligence collaboration has become a significant trend in library development,offering a new perspective on librarian professional identity research. STARA awareness reflects employees’ perceptions of the influence of smart technology,artificial intelligence,robotics,and algorithms on their future career development. This awareness helps librarians confront role transformation and advance human-intelligence collaboration within the library sector. Using a mixed-methods approach combining quantitative and qualitative analyses,we obtained 182 valid responses from librarians via a questionnaire survey;conducted semi-structured,in-depth interviews with 25 participants;and performed data coding and analysis based on grounded theory. The study presents a model examining the formation of STARA awareness and its impact on librarian professional identity. The model comprises seven core dimensions：STARA technological attributes,librarians’ use of STARA technologies,STARA applications,perceived impacts of STARA technologies,STARA awareness,librarians’ basic characteristics,and professional transformation and development. The findings demonstrate that the formation of STARA awareness is a long-term,multidimensional process. Current STARA awareness is characterized by the coexistence of opportunities and threats. Furthermore,STARA awareness positively reinforces and transformatively reconstructs librarians’ professional identities.
-
LIBRARY TRIBUNE. 2026, 46(4): 26-35.As the library sector diversifies its talent recruitment,a growing number of individuals without formal library and information science training are entering the field and becoming an important part of the librarian workforce. Due to their lack of systematic education in library and information science,these librarians often face more complex challenges when developing their professional identities. This paper focuses on the identity construction of non-professional librarians in the context of "career transition",and presents a conceptual model based on in-depth interviews with 30 non-professional librarians and grounded theory. The study indicates that their identity crisis stems from a disruption in professional cognitive continuity - a significant gap exists between their anticipated identity before transitioning and their actual perception after employment. This deepening confusion about their values exacerbates their disorientation and unease regarding their professional identity,thereby triggering identity deconstruction. However,with self-directed practices and support from the external environment,non-professional librarians can gradually overcome their identity crises. They can then redefine their professional identities,and achieve value identification,professional competence,role anchoring and meaning construction in their occupational practices. This marks the completion of identity construction,as individuals transition from "marginalization" to "subjectivity",forming a professional identity that is critical and adaptive.
-
LIBRARY TRIBUNE. 2026, 46(4): 36-50.Cultivating children’s reading habits from an early age is a vital mission for public libraries. The child-friendly concept and philosophy of children’s services in public libraries are based on the shared ideology of safeguarding children’s rights. These shared values and complementary practices have given rise to the innovative concept of the child-friendly library. Using a three-dimensional analytical framework of “policy tools--child-friendly library elements--policy participants”,this paper examines institutional manifestations of policies related to child-friendly libraries. Through a review of representative cases at home and abroad,the study analyzes innovative practices in various contexts,as well as the practical challenges of developing child-friendly libraries with regard to service accessibility,quality,quantity,and responsiveness to children’s needs. The study proposes that municipal libraries serve as central hubs guided by the child-friendly library concept to improve the quality and efficiency of public library services for children in metropolitan areas.
-
LIBRARY TRIBUNE. 2026, 46(4): 51-62.As prompts become the primary interface for human-generative AI interaction,the importance of prompt ethics as an emerging ethical issue in AI increases. However,prompt ethics is not merely an extension of traditional upstream ethics concerning data,algorithms,and computing power,nor is it an extension of downstream platform ethics. Instead,it serves as a midstream hub that connects these two domains,dynamically triggering and posing risks. This paper presents a two-dimensional analytical framework based on user intention and risk sources. The framework reveals two types of prompt-related ethical risks. One type involves security risks caused by malicious prompts that exploit model vulnerabilities or abuse its capabilities （e.g. prompt injection,jailbreaking,and manipulation）. The other type encompasses more universal risks inadvertently triggered by vast numbers of ordinary users during routine interactions （e.g. bias entrenchment and unintentional privacy leaks）. Addressing this complex risk landscape requires an integrated governance system encompassing technical safeguards represented by input filtering and adversarial training;organizational regulations centered on the formulation of standardized operating procedures and ethical guidelines;and societal collaboration involving legal frameworks,public prompt literacy enhancement,and platform governance. The goal of prompt ethics governance is not to limit technology,but to ensure the safe,fair,and effective release of generative AI's social value.
-
LIBRARY TRIBUNE. 2026, 46(4): 63-73.This paper proposes a collaborative framework driven by retrieval-augmented generation （RAG） to enhance the performance of large language models （LLMs） on multilingual aspect-based sentiment analysis （MABSA） tasks. First,the framework trains task-specific models on multilingual datasets to generate a dynamic multilingual knowledge base. Then,guided by RAG principles,the framework screens the knowledge base to identify the most relevant knowledge fragments for the current inputs;and introduces a chain-of-thought （CoT） mechanism to optimize reasoning pathways. This ultimately achieves collaborative inference between LLMs and task-specific models at the knowledge level,providing an interpretable solution for the task. Experimental results demonstrate that RAG-enhanced LLMs significantly outperform traditional LLMs on MABSA tasks,validating the effectiveness of this approach.
-
LIBRARY TRIBUNE. 2026, 46(4): 74-91.From a multi-source data perspective,this paper introduces a frontier topic mining/identification indicator system as well as a BERT-LDA model,based on academic papers,research grants,and patents. The study suggests that the system effectively identifies frontier topics within a research domain,and provides guidance for the data-driven quantitative identification of cutting-edge themes. Frontier topics in artificial intelligence span seven categories：machine learning and deep learning,natural language processing,computer vision,robotics,constraint and satisfaction problems,hardware and software,and applications. Frontier topics in various types of scientific and technological literature exhibit continuity and dynamic change during different time frames,with different types of literature emphasizing different frontier topics within the same period. In the field of artificial intelligence,frontier topics derived from different data sources exhibit stark evolutionary differences;and paper-based frontier topics vary considerably in evolutionary capacity across stages,with increasingly diverse developmental trajectories. Meanwhile,frontier topics reflected in research grants demonstrate pronounced evolutionary characteristics and diversification of development directions. Frontier topics reflected in patents show relatively minor differences during the evolutionary process,and their development directions are relatively stable. Significant differences exist between hotspots and emerging frontier topics emphasized at different stages throughout the evolutionary process.
-
LIBRARY TRIBUNE. 2026, 46(4): 92-104.Through a large-scale literature analysis,this paper explores the application scenarios and transformation paths of traditional library services empowered by generative artificial intelligence （GAI）. The study uses systematic review and meta-analysis （PRISMA） methods to identify relevant literature. Based on the core literature,it analyzes the key elements necessary to empower traditional library services with GAI. Three typical scenarios were identified：intelligent upgrading,intelligent transformation,and intelligent innovation. Accordingly,the study outlines the essential conditions for effectively deploying GAI in traditional library services and discusses strategies for risk prevention.
-
LIBRARY TRIBUNE. 2026, 46(4): 105-116.Immersive technology reduces cognitive load during the acquisition and comprehension of information by employing intuitive,vivid,and imaginative presentation methods. It provides innovative solutions to the challenges of disseminating medical knowledge and offers new perspectives on how medical libraries can navigate functional transformation and service innovation. In this study,we examined the immersive technology application developed by the Fudan University Medical Library since 2017. First,we invited 12 staff members to participate in semi-structured interviews. On-site observations captured user experiences and feedback. Subsequently,we conducted a coding analysis using the critical incident technique. Ultimately,our study revealed the steps involved in adopting and promoting immersive technologies in medical libraries to provide sustainable information services within specialized medical contexts. We analyzed the enabling characteristics of immersive technologies for medical libraries from an affordance perspective. Furthermore,we discussed core aspects at the managerial,environmental,and operational levels through a complementary assets lens,clarifying the information activities and sustainability challenges involved in advancing the effective implementation of smart services in medical libraries. Our goal was to provide clear,actionable guidance to help other libraries introduce new technologies,innovate information service models,and achieve functional transformation.
-
LIBRARY TRIBUNE. 2026, 46(4): 117-126.In response to competitive pressures and user retention challenges faced by university library document delivery services in the digital-intelligence era,this paper explores mechanisms that influence continued usage intentions from the perspective of user-perceived value,aiming to provide a theoretical basis and practical pathways for service innovation. The study presents a three-dimensional evaluation system that incorporates "functional-emotional-social" aspects by integrating customer-perceived value theory with the ECM-ISC model. The system’s dimensional connotations were contextualized within the document delivery setting using the PERVAL scale. A questionnaire survey of 288 university users in Beijing was conducted to verify the pathways through which perceived value affects satisfaction and continued usage intention using structural equation modeling. The findings reveal that social value exerts a total effect of 0.684 on continued usage intention in academic research contexts through both direct （β=0.299） and indirect （β=0.385） paths,confirming its role as a crucial link for users in building academic capital. Emotional value demonstrates a significant independent driving effect （β=0.343）,while functional value requires full mediation through satisfaction. Additionally,users from arts disciplines exhibited significantly lower satisfaction levels. Thus,the study proposes a "four-dimensional strategy enabled by digital intelligence" that aligns closely with the empirical evidence,including a resource collaboration chain （metadata interoperability）,an intelligent service chain （AI and blockchain）,a tiered response system,and a user experience closed loop （discipline-specific enhancement）. This provides a theoretically grounded and precisely targeted empirical paradigm for the digital-intelligent transformation of document delivery services in university libraries.
-
LIBRARY TRIBUNE. 2026, 46(4): 127-137.Focusing on AI-generated picture books,this paper employs a mixed experimental design and eye-tracking technology to explore the influence mechanism of three factors—prompt style,visual style,and color tone—on the reading experience of preschool children. A three-factor （2×2×2） experimental design was adopted,and multidimensional analyses were conducted through eye-tracking experiments and interviews. The results show that warm-toned picture books offer a better reading experience than cool-toned ones; and that image-descriptive prompts are more effective than story-descriptive ones. There is an interaction effect between visual style and prompt style. Story-descriptive prompts in a crayon-drawing style enhance the experience by providing more details （namely,detail compensation）,whereas image-descriptive prompts minimize differences in visual style. Eye-tracking data indicate that children pay more attention to character-related areas of interest than text-only ones,demonstrating an image-prioritized cognitive characteristic. Based on these findings,the study proposes strategies for generating AI picture books across five dimensions.
-
LIBRARY TRIBUNE. 2026, 46(4): 138-149.Exploring college students’ willingness to use AIGC tools for health information searches and the factors influencing this behavior is conducive to promoting the application of AIGC technology in health information services. A total of 24 college students were recruited through purposive sampling for semi-structured,in-depth interviews to understand their experiences and attitudes toward using AIGC tools for health information searches. The interview data were coded using a combined approach of grounded theory and content analysis. Based on SOR theory,a model of influence factors on the willingness to seek health information using AIGC tools was developed. The study suggests that college students hold a relatively positive intention to use AIGC tools to search for health information. The results reveal the pathways and mechanisms through which external （e.g.,AIGC tools and environmental features） and internal （e.g.,users） stimuli influence willingness via cognitive and emotional mediators. The study offers recommendations for enhancing health information literacy,optimizing AIGC tools,and improving health information services.
-
LIBRARY TRIBUNE. 2026, 46(4): 150-160.Examining the impact of digital burden on users’ information interaction behaviors on public digital cultural service platforms through generational differences can provide a theoretical and practical foundations for optimizing platform equity and bridging the digital divide. This study develops a digital burden scale for users of such platforms,and employs partial least squares structural equation modeling （PLS-SEM） to perform a path analysis on models A （non-elderly group,n=299） and B （elderly group,n=231）. The findings indicate that digital burdens on public digital cultural service platforms consist primarily of technical,emotional,physical,and temporal burdens,with elderly users experiencing significantly higher burdens than non-elderly users. Path analysis demonstrates that information interaction intention positively influences user behavior in both groups. However,there are generational differences in how digital burdens affect interaction intentions and behavior：elderly users are more likely than non-elderly users to reduce information exchange on the platforms due to digital burden. This study underscores the pivotal role of generational differences in the relationship between digital burden and information interaction behaviors,and provides targeted recommendations.
