Search Results (305)

Corporate reputation teams rely on media monitoring and qualitative research, both limited in speed and coverage when digital narratives form rapidly. This paper proposes SAPIENT (Sentinel-Augmented Population Intelligence for Emerging Narrative Tracking), a multi-agent system that links a sentinel layer over public text streams with a simulation layer that runs moderated, repeatable in silico focus-group sessions. The sentinel layer ingests social media, news, and forum text to produce a compact signal state (topics, sentiment, anomaly scores, risk labels), which conditions the simulation layer through an orchestrator. Persona agents and a moderator follow an Agentic Focus Group (AFG) protocol with repeated runs, variance reporting, and human review gates. We describe four sustainability communication scenarios: greenwashing backlash prediction, greenhushing risk assessment, campaign pre-testing, and crisis communication simulation. Nine experiments span 280 AFG runs across 20 conditions, three LLM backends (Claude Sonnet 4, GPT-4o, and Gemini 2.5 Flash), and a preregistered pilot human validation study with 54 participants. Signal conditioning improved simulation specificity ($p=0.012$). Cross-lingual sessions revealed a sentiment asymmetry between English and Turkish ($p=0.001$) with preserved persona rank ordering ($r=0.81$, $p=0.015$). Cross-model comparison showed consistent persona differentiation across all three backends (Pearson $r>0.92$, $p<0.002$ for all pairs). Sentiment was robust to prompt paraphrasing ($p=0.061$, n.s.), though credibility was sensitive to prompt wording ($p<0.001$). All significant results from Experiments 1–8 survived Benjamini–Hochberg correction. A preregistered pilot with 54 human participants on Prolific replicated the predicted credibility ranking across framing variants ($p=0.004$) but not the sentiment ranking, identifying a specific calibration target for future work. Full article

In multi-turn conversational AI, individually innocuous personally identifiable information (PII) fragments disclosed across successive turns can accumulate into a re-identification risk that no single utterance reveals on its own. Existing PII detectors operate on isolated utterances and therefore cannot track this cross-turn evidence build-up. We propose a stateful middleware guardrail whose core design principle is speaker-attributed entity isolation: every extracted PII fragment is attributed to its originating conversational participant, and evidence is accumulated in entity-isolated subgraphs that prevent cross-entity contamination. The system signals re-identification onset $t_{\mathrm{pred}}$ at the earliest turn where combination-based rules grounded in the uniqueness literature are satisfied. On a 184-record template-synthetic evaluation corpus, the gated NER configuration leads on primary timeliness (OW@5 $= 73.4\%$, MAE$= 1.357$ turns); the full system achieves OW@5 $= 70.7\%$ with MAE $= 2.442$ turns as an alternative operating mode for ambiguity-sensitive disclosure patterns. We further evaluate behavior on a 300-record mutation stress set, test RULE_B on the ABCD external corpus, and supplement RULE_A evaluation with both a proxy-labeled transfer analysis on PersonaChat and a manual annotation study on 151 Switchboard dialogues. The reported results should be interpreted as an initial empirical reference point rather than a sufficient endpoint for autonomous runtime enforcement. Full article

Background/Objectives: Since 2023, a significant increase in measles cases has been reported worldwide, and Italy has been among the most affected European countries. In this context, the integration of laboratory and epidemiological data enables timely case classification and helps distinguish between imported and indigenous cases, supporting disease control. However, most studies address only selected aspects of surveillance. Therefore, this study aimed to provide an integrated analysis of virological and epidemiological surveillance activities conducted between November 2023 and December 2025 by the Regional Reference Laboratory in the Emilia-Romagna Region (ERR). Methods: A total of 806 clinical samples (269 urine, 267 oral fluids—saliva or oropharyngeal swabs—and 270 sera) from 291 suspected measles cases were tested by molecular and/or serological methods, and MV genotyping was performed. Samples from discarded cases were also analysed for parvovirus B19 (B19V), human herpesvirus 6 (HHV-6), enterovirus (EV), and varicella zoster virus (VZV), chikungunya virus (CHIKV) and dengue virus (DENV). Results: Of 291 suspected cases, 176 (60.5%) were confirmed. Median age was 33 years, with 46% in the 15–39 year group. Vaccination status was available for 165: 90.3% were unvaccinated, 5.4% had one dose, and 4.2% had two doses. Notably, over half of confirmed cases occurred in areas with vaccine-hesitant communities. MV strain characterisation was performed in 99.4% of MV-RNA positive cases, with 84.3% genotype D8 and 15.6% genotype B3; 83% of strains were of indigenous origin, suggesting an ongoing endemic circulation. Clinical data showed complications in 19.3%, mainly pneumonia and diarrhoea. Additionally, differential diagnosis enabled the identification of the etiological agent in 37.5% of measles/rubella discarded cases, and 37.6% (29/77) tested positive for B19V. Conclusions: The study results highlight that effective measles surveillance must be supported by integrating timely virological diagnosis, molecular and epidemiological investigations, and differential diagnosis, to achieve the WHO goals of eliminating measles transmission. Full article

Evaluating emotional understanding in Large Language Models (LLMs) is challenging because assessments are subjective, ambiguous, multidimensional, and sensitive to controllable generation parameters. We developed a unified mathematical framework for characterizing LLM “emotional individuality” that integrates softmax sampling–temperature control (the decoding-time temperature parameter exposed by the API and typically used to modulate output randomness during token generation), fuzzy set theory with Shannon-type fuzzy entropy, and persona-based cognitive diversity analysis. We evaluated 36 API-accessible LLMs from seven major vendors on Japanese literary texts, using four personas each assigned a sampling temperature ($T\in \{0.1,0.4,0.7,0.9\}$), yielding 4227/4320 trial responses (97.8% coverage), of which 4067/4227 contained valid numeric emotion scores (96.2%). Temperature controllability varied approximately 25-fold (${\kappa }_M\in [0.039,0.982]$) with both positive and negative temperature–variance relationships across models. Because each sampling temperature is deterministically assigned to a persona in our design, ${\kappa }_M$ should be interpreted as an operational temperature–variance association across persona conditions rather than an isolated causal temperature effect. The model-level mean fuzzy entropy ranged from approximately 0.40 to 0.66, and the numerical stability consistency scores ranged from approximately 0.548 to 0.780. We also observed text-dependent structure, including genre-specific variation in the Interest–Sadness relationship. For practitioners, the framework is most directly useful as a benchmark-design and model-screening template for structured emotion-scoring tasks; its empirical conclusions remain limited to the present Japanese literary, text-only setting. Full article

This exploratory study examines how journalists in/coordinating investigations use a chatbot designed to reduce power asymmetries during remote work. Twelve freelancers across Africa, Europe, and India tested Obruche, a chatbot advisor covering risk mitigation, pay equality, tension de-escalation, and intellectual property protection. Drawing on the Unified Theory of Acceptance and Use of Technology, semi-structured interviews were coded for Performance Expectancy, Effort Expectancy, Facilitating Conditions, and Social Influence. Results show journalists gravitate towards chatbots that are cognisant of their location-specific challenges and able to provide information that facilitates access to media outlets or peers for future collaborations. Next-best-action responses that expanded user queries or offered role-play scenarios also left journalists feeling supported, less lonely, and not judged. However, the chatbot’s female persona, scepticism of artificial intelligence, and chatbot novelty may reduce user acceptance. Obruche’s potential areas of intervention are linked to eight types of organisational power. The chatbot mainly assisted journalists to confront or rebalance Control of Knowledge and Information, and Control of Scarce Resources, aiding users’ Ability to Cope with Uncertainty. This research contributes to recent qualitative studies on journalists’ well-being by demonstrating how chatbots can mitigate power imbalances between dispersed teams of journalists. The benefits and concerns presented may inform future designs of similar team-mediation chatbots. Full article

Large language models (LLMs) have shown strong performance in open-domain dialogue, yet they continue to struggle with long-term multi-session conversations (MSC), particularly in non-English languages such as Korean. In this work, we present a comprehensive empirical study on enhancing Korean MSC capabilities of LLMs through dataset construction, memory modeling, and parameter-efficient fine-tuning. We introduce an extended Korean MSC dataset that explicitly distinguishes between persona memory (long-term user attributes) and episode memory (short-term, event-driven information), enabling more effective memory management across sessions. Using this dataset, we evaluate LLM performance on three core MSC tasks: session summarization, memory update, and response generation. Our experiments reveal that Korean MSC is intrinsically more challenging than English MSC and that memory update and response generation require substantial reasoning ability. To address these challenges, we compare LoRA, DPO, MoE, CPT, Layer Tuning, and neuron-level tuning methods. Results consistently show that neuron tuning, guided by a novel language-specific neuron identification method based on activation scores and entropy, achieves superior performance and robustness, particularly in continual learning settings. Overall, our findings highlight neuron-level adaptation as an effective and interpretable approach for improving long-term conversational ability in low-resource languages. Full article

The rapid proliferation of large language models has created significant opportunities for personalized education, yet existing systems rarely account for user competency as a determinant of interaction quality. This study introduces Persona in The Loop (PITL), a dual-mode adaptive framework that recommends AI personas for blockchain and smart contract education applications. PITL employs 100 AI personas organized across two domains, ten sub-specialties, and five Dreyfus competency levels, recommending personas via either similarity-based mode grounded in Cognitive Load Theory or complementary mode grounded in the Zone of Proximal Development, with an adaptive switching mechanism driven by NASA-TLX cognitive load feedback. A mixed-method study with 150 participants using a 2 × 5 factorial design showed that the complementary mode produced higher learning gains, while the similarity-based mode yielded lower cognitive load and higher code quality. The adaptive mechanism outperformed both fixed-mode conditions on learning gain and code quality. The Mode × Dreyfus interaction was significant for cognitive load and task duration but not for learning gains, suggesting mode effects on learning outcomes are consistent across competency levels. Qualitative interviews with 20 participants corroborated quantitative findings. PITL offers a theoretically grounded and empirically validated approach to competency-based AI persona recommendation in educational contexts. Full article

Social media platforms have become primary communication channels for technical European universities. However, the extent to which global platform algorithms homogenize individual preferences across cultures remains underexplored. Although the current literature offers insights into the topic, none of the works consider the cross-national and multimodal nature of the phenomenon. In the current paper, we introduce the Media Information Processing System (MIPS), a privacy-preserving multimodal deep learning (DL) framework that incorporates large language models (LLMs), computer vision (CV), and knowledge graphs. We analyze data from 15,520 public videos shared by 2359 followers of six top technical universities from Romania, Germany, Italy, and Russia. The results of the study suggest that the degree of homogeneity of the followers’ interest profiles is markedly high. Statistical profiling of the data indicates that the interest profiles of the followers from different countries are positively correlated with a high degree of strength (mean Pearson r = 0.96; p > 0.90). Consensus clustering of the data reveals the existence of stable clusters of themes with high stability scores (>0.75), such as “Human Interaction Dynamics”. The results of the study contradict the traditional theory of regional cultural differentiation. Instead, the results suggest the existence of a new “digital student persona” that is characteristic of the academic lifestyle of students from different countries. Full article

This study is part of the FAIST research project—Agile, Intelligent, Sustainable and Technological Factory, coordinated by the Footwear Technology Centre of Portugal (CTCP), which aims to develop an innovative production process through the creation of a sustainable footwear model fully adapted to the user’s foot anatomy and personalized according to individual aesthetic preferences. Within this context, the need emerged to design an online platform with an interface capable of effectively addressing user needs throughout all stages of the personalization process, from the foot scanning to the aesthetic personalization of the model, while ensuring an efficient, intuitive, and pleasant navigation experience. Thus, this work aims to demonstrate how the design process of a footwear personalization platform, across its different phases, can contribute to the revitalization of the Portuguese footwear industry, as well as to describe its effectiveness, with the goal of being potentially adapted and implemented in similar contexts. The adopted methodology was based on the principles of Design Thinking, an approach centered on user needs. The development of the platform involved the creation of personas, the definition of the information architecture, the development of wireframes and workflows, and the execution of usability tests using the System Usability Scale (SUS). The results demonstrate a high success rate, validating the proposed solution with users and confirming the suitability of the applied methodologies. Full article

Large language models (LLMs) show promise for grading open-ended assessments but still exhibit inconsistent accuracy, systematic biases, and limited reliability across assignments. To address these concerns, we introduce SURE (Selective Uncertainty-based Re-Evaluation), a human-in-the-loop pipeline that combines repeated LLM prompting, uncertainty-based flagging, and selective human regrading. Three LLMs—gpt-4.1-nano, gpt-5-nano, and the open-source gpt-oss-20b—graded answers of 46 students to 130 open questions and coding exercises across five assignments. Each student answer was scored 20 times to derive majority-voted predictions and self-consistency-based certainty estimates. We simulated human regrading by flagging low-certainty cases and replacing them with scores from four human graders. We used the first assignment as a training set for tuning certainty thresholds and to explore LLM output diversification via sampling parameters, rubric shuffling, varied personas, multilingual prompts, and post hoc ensembles. We then evaluated the effectiveness and efficiency of SURE on the other four assignments using a fixed certainty threshold. Across assignments, fully automated grading with a single prompt resulted in substantial underscoring, and majority-voting based on 20 prompts improved but did not eliminate this bias. Low certainty (i.e., high output diversity) was diagnostic of incorrect LLM scores, enabling targeted human regrading that improved grading accuracy while reducing manual grading time by 40–90%. Aggregating responses from all three LLMs in an ensemble improved certainty-based flagging and most consistently approached human-level accuracy, with 70–90% of the grades students would receive falling inside human-grader ranges. A reanalysis based on outputs from a more diversified LLM ensemble comprised of gpt-5, codestral-25.01, and llama-3.3-70b-instruct replicated these findings but also suggested that large reasoning models such as gpt-5 might eliminate the need for human oversight of LLM grading entirely. These findings demonstrate that self-consistency-based uncertainty estimation and selective human oversight can substantially improve the reliability and efficiency of AI-assisted grading. Full article

The distinction between signs and expressions is essential to unlock Deleuze’s interpretation of Spinoza. However, during a lecture delivered on 13 January 1981, Deleuze makes a passing remark that complicates this distinction. For Spinoza, Christ’s religion, like political society, is a systems of signs pertaining to the collective imagination that nevertheless is meant to facilitate the transition towards the domain of expressions, that is, to the domain of reason and philosophy. The aim of this paper is to shed light on this ambiguity between signs and expressions in Deleuze’s work on Spinoza. First, I discuss the scattered passages in Spinoza’s oeuvre dealing with the figure of Christ. I then go on to reconstruct Deleuze’s Spinozistic taxonomy of signs. Third, I reconstruct Deleuze’s comparison between Spinoza and Hobbes regarding the emergence of political society from the state of nature. I then propose a close reading of chapter 7 of the Theological-Political Treatise to argue that Christ’s religion, according to Spinoza, should be seen as fulfilling the function of political society in times of crisis. I end with an extensive analysis of Spinoza’s formula “the Spirit of Christ, that is, the idea of God” in light of Deleuze’s reading of the first half of Ethics V. To conclude, I suggest we look at Christ as the conceptual persona of Spinozism. Full article

This essay offers a philosophical and bioethical upholding of Dignitas Personae §27, which cautions against the use of human gene editing (HGE) for non-therapeutic purposes. After situating the debate within the historical development of gene-editing technologies, the essay classifies enhancement-oriented interventions—physical, behavioral, and cognitive—and argues that such practices risk violating human dignity, diminishing authentic freedom, and promoting a deterministic anthropology. Drawing on a personalist framework, the analysis incorporates insights from neuroscience, genetics, and natural law. In the second part, the essay examines Aristotelian–Thomistic metaphysics, integrating Ernest Mayr’s notion of teleonomy to explain how the rational soul actualizes its perfect operations. It is argued that non-therapeutic HGE, especially germline modifications, may disrupt the ontological structure of the human person by impairing the soul’s expression through properly disposed prime matter. Ultimately, Dignitas Personae stands as a coherent and prescient response to emerging biotechnologies, defending the human person against technocratic reductionism and the ideological drive to transcend our embodied finitude. Full article

Accessing and understanding medication information can be challenging for many people, especially when patient information leaflets (PILs) are long, complex, and printed in small font. This study presents MediChat, an open-source, locally executable chatbot designed to provide reliable, easy-to-read answers to medication-related questions based exclusively on official PILs. MediChat follows a retrieval-augmented generation (RAG) architecture: PILs from the Austrian Medicinal Product Index are received via API, converted to text, split into overlapping chunks, embedded, and stored in a Chroma vector database. From there the top-k relevant chunks are retrieved, and Llama 3.1 generates German responses based on this evidence. The system was evaluated using a hybrid framework. Quantitatively, 200 yes/no questions across ten drugs were answered with 80% accuracy, overall precision 0.977, recall 0.686, F1-score 0.806, and a mean response time of 727 ms. Qualitatively, two personas were used in eight simulated dialogues. Response times were around 1.1–1.3 s, and task completion exceeded 85% with high ratings for relevance and quantity. These results indicate that an open-source RAG chatbot can deliver leaflet-grounded, user-friendly medication information and provide a reproducible template for future healthcare chatbot evaluations. Full article

Background/Objectives: Adoption of user-centred design methods is essential in healthcare applications because it ensures that complex workflows are shaped around real users’ needs and behaviours, improving usability, accessibility, and sustainability. The use of user-centred design in healthcare applications still presents open challenges for identifying user requirements, including diverse stakeholder needs, limited user availability, complex interaction workflows, and organizational constraints. To address these challenges, this paper proposes a user-centred interaction design framework that systematically supports the identification and translation of user needs into actionable design requirements. Methods: The framework integrates user-centred design principles with generative tools, employing the Persona-and-Scenario method to transform user insights into actionable design requirements. By actively involving healthcare stakeholders, the framework ensures that both explicit and latent needs are captured. Results: The framework was implemented through two co-design events, which provided valuable feedback on data collection, visualization, interaction modalities, and privacy considerations. These insights were translated into functional, usability, and interface requirements for the Change Management Platform (CMP) for the KEEPCARING project. Conclusions: This framework introduces a structured, scenario-driven process that actively engages stakeholders in envisioning future states rather than merely refining existing systems. Its application demonstrates promising indications that it enhances requirement elicitation, promotes cross-stakeholder alignment, and yields higher-quality, contextually relevant design requirements. Full article

Social media has become a vital source for humanitarian organizations to gather information during crises. However, existing multimodal classification methods operate primarily as isolated systems, while neglecting external references crucial for accurate judgment. Furthermore, while user comments can provide valuable context, they are often scarce during the early stages of a crisis. To address these limitations, we propose a framework named Mix-Persona Comment Generation with Geographically Enhanced Context Retrieval for LLM Instruction Fine-tuning (MPCG-GECR). To mitigate comment scarcity, we employ a Synthetic Persona Generator (SPG) that prompts LLMs to adopt diverse mix-personas, generating synthetic comments that simulate multi-perspective public discourse. To incorporate external references, we introduce a Geographically Enhanced Context Retrieval (GECR) module. Unlike standard retrieval approaches, GECR utilizes a hybrid re-ranking strategy to identify samples that are both multimodally similar and geographically consistent, serving as reliable reference anchors for the LLM. By integrating these social perspectives and geographic references into a unified instruction-tuning format, we transform the classification task into a context-aware text generation problem and fine-tune the LLM using Low-Rank Adaptation (LoRA). Extensive experiments on the CrisisMMD and DMD datasets demonstrate that MPCG-GECR effectively overcomes data scarcity and context isolation, significantly outperforming existing methods. Full article