Carnegie Mellon University Qatar

Abstract We highlight fairness as the most relevant moral belief system and examine how views on fairness interact with animal familiarity and affinity to predict support for conservation. Our study includes respondents from six cultural groups: the Arabian Gulf, Middle East and North Africa, Sub-Saharan Africa, South Asia, Southeast Asia, and so-called WEIRD (Western, Educated, Industrialized, Rich, Democratic) participants. The results reveal interactions between fairness and animal familiarity, as well as fairness and animal affinity. Across all animals surveyed, after accounting for cultural differences, support for animal conservation rises among individuals who believe animals are mistreated in their natural environments, even if they dislike or are unfamiliar with the animal, compared to those who do not perceive such unfairness. This suggests that perceptions of unfair treatment are a key factor in increasing conservation support, particularly among individuals who have less affinity for or knowledge about the animal. Therefore, emphasizing the unfair treatment of animals may be an effective messaging strategy to foster positive attitudes toward animal conservation.

Insider threats remain a persistent challenge in cybersecurity due to the deceptive nature of malicious activities conducted under legitimate user accounts. This paper presents a session-based detection framework integrating SMOTE-IPF oversampling and XGBoost classification to address temporal context limitations and class imbalance in insider threat datasets. To mitigate the extreme class imbalance characteristic of insider threat datasets, the framework integrates SMOTE-IPF, an advanced oversampling technique that maintains minority class structure while reducing overfitting. The model, trained with a GPU-accelerated XGBoost classifier, achieves notable performance improvements: $50 \%$ recall for threat instances at the F1-optimal threshold, 99.995 % accuracy for normal activity, and only four false positives. An ROC-AUC of 0.9429 and an F1score of 0.5263 demonstrate the model’s effectiveness in balancing precision and recall. These results indicate the proposed approach can enhance threat identification while maintaining operational feasibility in high-stakes security environments.

One way to cope with crises is by attributing their ultimate causes to malevolent conspiracies. As crises are rarely simple, and may involve an interplay between multiple, co-occurring threats, we suggest that conspiracy thinking mainly occurs among individuals who experience conditions of threat complexity – such as socioeconomic vulnerability paired with a sense of helplessness in society, and who are also sufficiently paranoid to infer a conspiracy. In the present study, we focused on financial strain and disempowerment, as two relevant threats which were both dramatically affected by the COVID-19 pandemic, and hypothesized a three-way interaction between financial strain, disempowerment and paranoia in predicting conspiracy thinking. This hypothesis was supported in both cross-national ( N = 64,130) and longitudinal data ( N = 11,159), collected during the COVID-19 pandemic. Implications of the results for understanding the tendency to reduce multiple threats to a single cause are discussed.

In the context of multi-omics data analytics for various diseases, transcriptome-wide association studies leveraging genetically predicted gene expression hold promise for identifying novel regions linked to complex traits. However, existing methods for multi-tissue gene expression prediction often fail to account for tissue-tissue expression interactions, limiting their accuracy and effectiveness. This research addresses the challenge of predicting gene expression across multiple tissues by incorporating tissue-tissue expression correlations based on a nonlinear multivariate model. Our findings demonstrate that this model excels in estimating tissue-tissue interactions and accurately predicting missing data. These results have significant implications for multi-omics data analytics and transcriptome-wide association studies, suggesting a novel approach for identifying regions associated with complex traits.

Machine unlearning has garnered widespread attention, due to various reasons, including privacy-preserving, model usability, and legal regulations. It requires model providers to unlearning users' data from models upon receiving unlearning request. Recommendation systems have also been extensively researched in the field of deep learning, particularly within the context of big data environments. However, little research can be found to verify the effectiveness of unlearning approach using pure tabular data-based recommendation scenario. In this paper, we propose a recommendation unlearning verification (RUV) scheme based on non-influential trigger data, which fills this gap. Users can use the recommendation rate for selected target items to determine whether the recommendation system complies with unlearning requests. Evaluation results on real datasets confirm the efficiency and effectiveness of our proposed RUV scheme.