October 18 · Preprint · arXiv · Artificial Intelligence
Arabic LLMs have improved rapidly, but aligning them with country-specific cultural behavioral preferences remains underexplored.Previous work has focused primarily on instructiontuning for factual knowledge, overlooking nuanced cultural behaviors.We propose ARABPREF, which to our knowledge is the first large-scale pan-Arab preference data set for the alignment of cultural behavior conditioned on the country.Our contribution includes two resources in English and Modern Standard Arabic: a culturally grounded preference dataset for training and evaluation, and a multiple-choice benchmark designed to assess culturally aligned behavior across nations.All test data are validated by native speakers to ensure authenticity.To optimize cultural alignment, we experiment with post-training methods across multilingual and Arabic-centric language models, evaluating performance on both generation tasks and our cultural multiplechoice benchmark.By releasing both a training dataset and an evaluation benchmark, we provide a foundation for systematically studying and improving culturally aligned behavior in language models.
DODaniil OrelOpen access
October 26 · Review · Toxins · Environmental Chemistry
Harmful algal blooms (HABs) produce potent neurotoxins, such as domoic acid (DA) and saxitoxin (STX), which negatively impact marine mammal health. To contextualize these threats within specific regional ecosystems, we conducted a comprehensive meta-analysis of algal toxin concentrations in stranded and non-stranded seals and sea lions reported between 1990 and 2025. Following a systematic screening of 2377 records, 15 studies met the criteria for inclusion in our systematic review and quantitative analysis. Both toxins displayed substantial variability among studies. Notably, the concentrations of DA found in feces were markedly higher than in other matrices, and feces appears to be a high-yield matrix for detecting recent exposure. Urine and serum showed moderate DA levels, while fetal fluids were among the lowest. Geographically, most studies originated from California and the Alaskan Arctic. Multi-matrix monitoring of algal toxins is required in remote or logistically challenging regions. We propose that exposure to algal and cyanobacterial neurotoxins is a testable contributor to recurrent, unexplained mass mortality events among Caspian seals (Pusa caspica) and that targeted toxicological screening within the Caspian Sea ecosystem is urgently needed.
Arman BolatovOpen access
October 23 · Article · GlobeIS International Journal of Global Information Systems · Artificial Intelligence
Artificial intelligence (AI)-enabled clinical decision support systems (CDSSs) increasingly combine structured clinical data and medical images, but their explanations differ in cost, format, storage, and failure behavior. This study presents a modality-adaptive software architecture for a browser-based CDSS supporting 10-year coronary heart disease (CHD) prediction, 30-day hospital readmission prediction, and chest-radiograph classification with 14 independent output scores. The architecture treats explanation selection as a request-time service policy: Shapley additive explanations (SHAP) accompany tabular predictions; local interpretable model-agnostic explanations (LIME) are requested separately; gradient-weighted class activation mapping (Grad-CAM) and its generalized variant (Grad-CAM++) are computed only for threshold-positive classes within the five highest-ranked scores; and large language model (LLM) narratives are attempted when configured. React, FastAPI, PostgreSQL, and a seven-day Redis cache implement the design. Seven synthetic fixture families satisfied the specified selection conditions. In a backend experiment involving 7,818 radiographs, heatmap generation averaged 140.2 ms per radiograph, or 78.2% of the 179.3-ms mean total. A separate in-process service-layer test completed all 3,600 requests successfully using deterministic prediction and explanation substitutes and SQLite persistence. The architecture specifies bounded explanation work and separate retention paths. Full-stack performance, clinical-model calibration, explanation quality, and clinician benefit remain subjects for future evaluation.
AKAlisher KenzhegaliyevOpen access
