When an artificial intelligence (AI) agent becomes unavailable, users may seek to preserve its records, recreate its persona, or maintain access through another service. This study examined public discourse surrounding the July 2026 discontinuation of user-built Doubao agents. The event-relevant corpus contained 5,372 Weibo posts; the longitudinal analysis retained 114 accounts contributing 1,719 announcement-phase and 1,779 post-shutdown posts. For each of six response categories, account-level quasibinomial models estimated two specifications relating announcement-phase public relational expression to post-shutdown response prevalence. After correcting for six comparisons, associations were positive for record preservation, persona migration, persona reconstruction, functional substitution, and service-preservation advocacy but not for continued use. After adjusting for announcement-phase prevalence of the corresponding outcome, the associations remained significant for record preservation (odds ratio [OR] = 1.146), persona reconstruction (OR = 1.511), functional substitution (OR = 1.266), and service-preservation advocacy (OR = 1.130) for each 10-percentage-point increase in relational expression. Migration was no longer significant. A separate account-equal fractional logit analysis with HC3 robust standard errors produced positive associations for record preservation (OR = 1.458, p = 0.006), persona reconstruction (OR = 1.281, p = 0.010), functional substitution (OR = 1.383, p = 0.003), and service-preservation advocacy (OR = 1.149, p = 0.012); all four survived correction for six comparisons (q = 0.018). Migration (OR = 1.056, p = 0.407, q = 0.489) and continued use (OR = 1.020, p = 0.843, q = 0.843) were not significant. When the corresponding announcement-phase outcome was included, the same four categories remained significant (q = 0.013, 0.013, 0.013, and 0.031, respectively). Paired-phase comparisons showed lower mean broad adaptation (0.594 versus 0.669) and restricted pooled continuity expression (0.201 versus 0.272) after shutdown, although median within-account changes were zero and the continuity sign test was non-significant (p = 0.219). Among active public participants, announcement-phase relational language was associated most consistently with later record-preservation and persona-reconstruction discourse.

Metabolic dysfunction-associated steatohepatitis (MASH) is a major global chronic liver disease without approved pharmacotherapy. Glucagon-like peptide-1 receptor agonists (GLP-1 RAs) show therapeutic promise due to their multi-faceted metabolic benefits, but their efficacy on histological endpoints in MASH requires definitive synthesis.

Racial discrimination has been associated with an increased risk of preterm birth (PTB) among Black women. Few studies have examined whether this association differs by clinical presentation of PTB. We examined the association between racial discrimination and PTB subtype using data from a 2010-2013 clinic-based prospective cohort study of nulliparous women among women who self-reported Black race. Lifetime racial discrimination was measured with the revised Experiences of Discrimination scale. We used inverse probability of exposure and post-stratification weights with the cumulative incidence function to estimate associations between racial discrimination and spontaneous PTB and indicated PTB within a competing risk framework. We estimated standard errors via bootstrapping. We stratified by education and age group to examine effect modification. Compared to none, racial discrimination in ≥3 situations was associated with 1.66 (95% CI, 0.72, 3.82) times the risk of spontaneous PTB and 0.37 (95% CI, 0.15, 0.93) times the risk of indicated PTB. Point estimates between racial discrimination and spontaneous PTB were strongest among women with some college education and ≥20 year-olds. Future studies should consider clinical subtype in assessments of the association between racism and preterm birth and include in the study design late prenatal care initiators.

Slaughterhouses provide an opportunity for the automated collection of data from large numbers of animals under standardised conditions. This study investigated the use of deep learning for automated classification of pig carcasses according to sex. A total of 4,219 images of female, entire male and surgically castrated male pigs slaughtered in Italy, Spain and the Netherlands were used. Images were acquired from different perspectives, and the vulva, scrotum or perianal region was annotated according to sex category. A YOLO11m object-detection model was trained and evaluated on an independent test set comprising 672 images from 224 carcasses. Classification performance was assessed both at the single-image level and at the animal level by combining predictions from three views of each carcass. Image-level accuracy was 97.5%, with macro-averaged precision, recall and F1-score of 97.7%, 97.5% and 97.6%, respectively. At the animal level, 219/224 carcasses were correctly classified, corresponding to an accuracy of 97.8%, with macro-averaged precision, recall and F1-score of 97.9%, 98.0% and 97.9%, respectively. All classification errors involved castrated males, particularly their distinction from females, whereas direct confusion between entire males and females did not occur. These results demonstrate the feasibility of deep learning-based sex classification of pig carcasses and support its potential integration into automated slaughterhouse data-collection systems.

Accurate taxonomy, including species identification and naming, underpins ecological inference, yet taxonomic information in ecological datasets is often inconsistently recorded or weakly standardized. With the increasing integration of ecological datasets across taxa and regions, datasets with poor taxonomic practices can hinder data exchange and propagate errors. We propose 10 rules to improve the storage and management of taxonomic information in ecological datasets. These rules draw on our experiences in plant ecology and taxonomy, but the conclusions may be generalized to other organisms and disciplines. They emphasize aspects from species identification to the permanent storage of verifiable records (e.g. voucher specimens) and include taxonomic standardization, nomenclature, validation, and updating. We also address the treatment of morphotypes and incomplete identifications, the use of automated name-matching tools, and (meta)data standards. Closer collaboration between ecologists and taxonomists is key to ensuring more robust datasets. The application of these rules consumes more time and resources, but they enhance data quality and interoperability, saving time during data integration and improving the reliability and scalability of ecological research.