Agora

The paper aims to examine the operation of student business incubators, which play an increasingly important role in fostering entrepreneurship, innovation, and university–industry collaboration, especially in emerging entrepreneurial ecosystems such as Croatia’s. The research was conducted through in-depth interviews with five student business incubator managers. The study identified the dominant services provided by incubators, the barriers faced by their tenants, incubator outcomes, different forms of cooperation with the business sector, and plans for the future. The paper concludes that the most successful incubators provide support in: (1) idea development; (2) prototyping; (3) investor search; (4) market entry; and (5) identification of opportunities based on the needs of the business sector. The results will help other student business incubators conduct benchmarking and assist organizations planning to establish a student business incubator in implementing the identified successful practices.

We propose OutboundEval, a comprehensive benchmark for evaluating large language models (LLMs) in expert-level intelligent outbound calling scenarios. Unlike existing methods that suffer from three key limitations - insufficient dataset diversity and category coverage, unrealistic user simulation, and inaccurate evaluation metrics - OutboundEval addresses these issues through a structured framework. First, we design a benchmark spanning six major business domains and 30 representative sub-scenarios, each with scenario-specific process decomposition, weighted scoring, and domain-adaptive metrics. Second, we develop a large-model-driven User Simulator that generates diverse, persona-rich virtual users with realistic behaviors, emotional variability, and communication styles, providing a controlled yet authentic testing environment. Third, we introduce a dynamic evaluation method that adapts to task variations, integrating automated and human-in-the-loop assessment to measure task execution accuracy, professional knowledge application, adaptability, and user experience quality. Experiments on 12 state-of-the-art LLMs reveal distinct trade-offs between expert-level task completion and interaction fluency, offering practical insights for building reliable, human-like outbound AI systems. OutboundEval establishes a practical, extensible, and domain-oriented standard for benchmarking LLMs in professional applications.

The intestinal microbiome is essential for gastrointestinal and overall health, yet its response to air pollution in children remains underexplored. In a study involving 412 young children from the ENVIRONAGE cohort, stool samples were analysed via Illumina Miseq sequencing to assess microbiome alpha diversity (observed richness, species evenness, and Shannon diversity) and composition. Exposure to previous year particulate air pollution (black carbon, PM2.5, coarse PM, and PM10) was modeled using high-resolution spatial-temporal interpolation models. Multiple linear regression models were adjusted for a priori selected covariables and stratified by sex. Furthermore, we performed a differential relative abundance analysis at family and genus level, while accounting for the same covariables. Statistically significant effect modification by sex was apparent for several intestinal alpha diversity indices and air pollutants. In boys, we observed negative associations between particulate air pollution exposure and intestinal microbiome richness (estimates ranging from -5.55 to -9.06 per interquartile range (IQR) increase in particulate air pollution exposure) and Shannon diversity (estimates ranging from -0.058 to -0.095 per IQR increase). Differently, in girls non-significant positive associations were observed with species evenness (estimates ranging from 0.019 to 0.020 per IQR increase) and Shannon diversity (estimate 0.065 per IQR increase in black carbon). After multiple testing correction, we reported several bacterial families and genera (Streptococcaceae, Clostridiales Incertae Sedis XIII, Coriobacteriaceae, Streptococcus, and Paraprevotella) to be oppositely associated with particulate air pollution exposure in boys and girls. Our findings show a sex-dependent association between particulate air pollution exposure and intestinal microbiome composition, highlighting boys as potentially more vulnerable to diversity loss associated with childhood exposure to particulate pollution.

Until recently, Legal and Forensic Linguistics were not prominent in France, despite William Labov's influential 1989 expert testimony in the US. The lag is due to the Germano-Roman legal system's lesser reliance on non-legal experts, its limited connections with academia, and the perception that linguistic expertise is unnecessary beyond the boundaries of law degrees and classical forensic fields. Since 2013, initiatives like the annual French-speaking summer school in Forensic Linguistics and collaborations with Universite Savoie Mont-Blanc and Universite du Quebec a Trois Rivieres have begun to integrate Forensic Linguistics into French academia. These efforts have led to the development of Forensic Linguistics, particularly in document analysis, and the establishment of the CY Forensic School, which partners internationally to advance forensic science education in France. This paper reviews the state of the field in France and examines two exemplary cases, those of Petit Gregory (1984 for the start) and AGRIF v. ZEP (2015), which showcase the emerging role of linguists in the legal arena.

Sickle cell disease (SCD) is a significant health burden in the Democratic Republic of the Congo (DRC). This study aims to identify predictive factors of mortality in SCD children admitted to emergency care in Lubumbashi, DRC. We performed a non -interventional cohort follow-up on SCD patients aged 0 to 16 admitted for a "true emergency". Demographic, clinical, and biological data were collected. Univariate and multivariate logistic regression analyses were performed to identify significant risk factors associated with mortality. Among the 121 patients included, 24 died during the follow-up period. Univariate regression revealed age, Mikobi score, referral origin, stroke, and severe infection as significant risk factors. Multivariate analyses identified Hb, WBC, SR, and LDH as predictive factors of mortality. Notably, patients aged 12 to 16 years faced a higher risk, shifting the age of mortality from early to late childhood and adolescence. This study provides valuable insights into mortality risk factors for paediatric SCD patients during acute crises. Early diagnosis, regular follow-up, and therapeutic education are essential to improve patient outcomes and survival rates. These findings contribute to better disease management and targeted interventions, aiming to reduce mortality associated with SCD.