By allowing models to predict without task-specific training, in-context learning (ICL) with pretrained LLMs has enormous potential in NLP. However, a number of problems persist in ICL. In particular, its performance is sensitive to the choice and order of in-context examples. Given the same set of in-context examples with different orderings, model performance may vary between near random to near state-of-the-art. In this work, we formulate in-context example ordering as an optimization problem. We examine three problem settings that differ in the assumptions they make about what is known about the task. Inspired by the idea of learning from label proportions, we propose two principles for in-context example ordering guided by model's probability predictions. We apply our proposed principles to thirteen text classification datasets and nine different autoregressive LLMs with 700M to 13B parameters. We demonstrate that our approach outperforms the baselines by improving the classification accuracy, reducing model miscalibration, and also by selecting better in-context examples.

Designing proper treatment plans to manage diabetes requires health practitioners to pay heed to the individuals remaining life along with the comorbidities affecting them. Older adults with Type 2 Diabetes Mellitus (T2DM) are prone to experience premature death or even hypoglycaemia. The structured dataset utilized has 68 potential mortality predictors for 275,190 diabetic U.S. military Veterans aged 65 years or older. A new target variable is invented by combining the two original target variables. Outliers are handled by discretizing the continuous variables. Categorical variables have been dummy encoded. Class balancing is achieved by random under-sampling. A benchmark regression model is built using Multinomial Logistic Regression with LASSO. Chi-Squared and Information Gain are the filter-based feature selection techniques utilized. Classifiers such as Multinomial Logistic Regression, Random Forest, Extreme Gradient Boosting (XGBoost), and One-vs-Rest classifier are employed to build various models. Contrary to expectations, all the models have constantly underperformed. XGBoost has given the highest accuracy of 53.03 percent with Chi-Squared feature selection. All the models have consistently shown an acceptable performance for Class 3 (remaining life is more than 10 years), significantly low for Class 1 (remaining life is up to 5 years), and the worst for Class 2 (remaining life is more than 5 but up to 10 years). Features analysis has deduced that almost all input variables are associated with multiple target classes. The high dimensionality of the input data after dummy encoding seems to have confused the models, leading to misclassifications. The approach taken in this study is ineffective in producing a high-performing predictive model but lays a foundation as this problem has never been viewed from a multiclass classification perspective.

Prediction of symptomatic crises in chronic diseases allows to take decisions before the symptoms occur, such as the intake of drugs to avoid the symptoms or the activation of medical alarms. The prediction horizon is in this case an important parameter in order to fulfill the pharmacokinetics of medications, or the time response of medical services. This paper presents a study about the prediction limits of a chronic disease with symptomatic crises: the migraine. For that purpose, this work develops a methodology to build predictive migraine models and to improve these predictions beyond the limits of the initial models. The maximum prediction horizon is analyzed, and its dependency on the selected features is studied. A strategy for model selection is proposed to tackle the trade off between conservative but robust predictive models, with respect to less accurate predictions with higher horizons. The obtained results show a prediction horizon close to 40 minutes, which is in the time range of the drug pharmacokinetics. Experiments have been performed in a realistic scenario where input data have been acquired in an ambulatory clinical study by the deployment of a non-intrusive Wireless Body Sensor Network. Our results provide an effective methodology for the selection of the future horizon in the development of prediction algorithms for diseases experiencing symptomatic crises.

Transformer based code models have impressive performance in many software engineering tasks. However, their effectiveness degrades when symbols are missing or not informative. The reason is that the model may not learn to pay attention to the right correlations/contexts without the help of symbols. We propose a new method to pre-train general code models when symbols are lacking. We observe that in such cases, programs degenerate to something written in a very primitive language. We hence propose to use program analysis to extract contexts a priori (instead of relying on symbols and masked language modeling as in vanilla models). We then leverage a novel attention masking method to only allow the model attending to these contexts, e.g., bi-directional program dependence transitive closures and token co-occurrences. In the meantime, the inherent self-attention mechanism is utilized to learn which of the allowed attentions are more important compared to others. To realize the idea, we enhance the vanilla tokenization and model architecture of a BERT model, construct and utilize attention masks, and introduce a new pre-training algorithm. We pre-train this BERT-like model from scratch, using a dataset of 26 million stripped binary functions with explicit program dependence information extracted by our tool. We apply the model in three downstream tasks: binary similarity, type inference, and malware family classification. Our pre-trained model can improve the SOTAs in these tasks from 53% to 64%, 49% to 60%, and 74% to 94%, respectively. It also substantially outperforms other general pre-training techniques of code understanding models.

Reaching consensus on a commonly accepted definition of AI Fairness has long been a central challenge in AI ethics and governance. There is a broad spectrum of views across society on what the concept of fairness means and how it should best be put to practice. In this workbook, we tackle this challenge by exploring how a context-based and society-centred approach to understanding AI Fairness can help project teams better identify, mitigate, and manage the many ways that unfair bias and discrimination can crop up across the AI project workflow. We begin by exploring how, despite the plurality of understandings about the meaning of fairness, priorities of equality and non-discrimination have come to constitute the broadly accepted core of its application as a practical principle. We focus on how these priorities manifest in the form of equal protection from direct and indirect discrimination and from discriminatory harassment. These elements form ethical and legal criteria based upon which instances of unfair bias and discrimination can be identified and mitigated across the AI project workflow. We then take a deeper dive into how the different contexts of the AI project lifecycle give rise to different fairness concerns. This allows us to identify several types of AI Fairness (Data Fairness, Application Fairness, Model Design and Development Fairness, Metric-Based Fairness, System Implementation Fairness, and Ecosystem Fairness) that form the basis of a multi-lens approach to bias identification, mitigation, and management. Building on this, we discuss how to put the principle of AI Fairness into practice across the AI project workflow through Bias Self-Assessment and Bias Risk Management as well as through the documentation of metric-based fairness criteria in a Fairness Position Statement.

The pervasive use of social media platforms, such as Facebook, Instagram, and X, has significantly amplified our electronic interconnectedness. Moreover, these platforms are now easily accessible from any location at any given time. However, the increased popularity of social media has also led to cyberbullying.It is imperative to address the need for finding, monitoring, and mitigating cyberbullying posts on social media platforms. Motivated by this necessity, we present this paper to contribute to developing an automated system for detecting binary labels of aggressive tweets.Our study has demonstrated remarkable performance compared to previous experiments on the same dataset. We employed the stacking ensemble machine learning method, utilizing four various feature extraction techniques to optimize performance within the stacking ensemble learning framework. Combining five machine learning algorithms,Decision Trees, Random Forest, Linear Support Vector Classification, Logistic Regression, and K-Nearest Neighbors into an ensemble method, we achieved superior results compared to traditional machine learning classifier models. The stacking classifier achieved a high accuracy rate of 94.00%, outperforming traditional machine learning models and surpassing the results of prior experiments that utilized the same dataset. NTRODUCTION Today, social networking sites play a significant role in our daily lives. We use social media for various communications, encompassing entertainment, education, personal development, and the workplace. The revolutionary nature of these platforms has made it much easier to connect with people across long distances [1]. Technological advancements have transformed the way we communicate, share information, and interact with communities globally [2].

This study assesses deep learning models for audio classification in a clinical setting with the constraint of small datasets reflecting real-world prospective data collection. We analyze CNNs, including DenseNet and ConvNeXt, alongside transformer models like ViT, SWIN, and AST, and compare them against pre-trained audio models such as YAMNet and VGGish. Our method highlights the benefits of pre-training on large datasets before fine-tuning on specific clinical data. We prospectively collected two first-of-their-kind patient audio datasets from stroke patients. We investigated various preprocessing techniques, finding that RGB and grayscale spectrogram transformations affect model performance differently based on the priors they learn from pre-training. Our findings indicate CNNs can match or exceed transformer models in small dataset contexts, with DenseNet-Contrastive and AST models showing notable performance. This study highlights the significance of incremental marginal gains through model selection, pre-training, and preprocessing in sound classification; this offers valuable insights for clinical diagnostics that rely on audio classification.

Online transaction fraud presents substantial challenges to businesses and consumers, risking significant financial losses. Conventional rule-based systems struggle to keep pace with evolving fraud tactics, leading to high false positive rates and missed detections. Machine learning techniques offer a promising solution by leveraging historical data to identify fraudulent patterns. This article explores using the personalised PageRank (PPR) algorithm to capture the social dynamics of fraud by analysing relationships between financial accounts. The primary objective is to compare the performance of traditional features with the addition of PPR in fraud detection models. Results indicate that integrating PPR enhances the model's predictive power, surpassing the baseline model. Additionally, the PPR feature provides unique and valuable information, evidenced by its high feature importance score. Feature stability analysis confirms consistent feature distributions across training and test datasets.

Digital pathology involves converting physical tissue slides into high-resolution Whole Slide Images (WSIs), which pathologists analyze for disease-affected tissues. However, large histology slides with numerous microscopic fields pose challenges for visual search. To aid pathologists, Computer Aided Diagnosis (CAD) systems offer visual assistance in efficiently examining WSIs and identifying diagnostically relevant regions. This paper presents a novel histopathological image analysis method employing Weakly Supervised Semantic Segmentation (WSSS) based on Capsule Networks, the first such application. The proposed model is evaluated using the Atlas of Digital Pathology (ADP) dataset and its performance is compared with other histopathological semantic segmentation methodologies. The findings underscore the potential of Capsule Networks in enhancing the precision and efficiency of histopathological image analysis. Experimental results show that the proposed model outperforms traditional methods in terms of accuracy and the mean Intersection-over-Union (mIoU) metric.

An important step towards explaining deep image classifiers lies in the identification of image regions that contribute to individual class scores in the model's output. However, doing this accurately is a difficult task due to the black-box nature of such networks. Most existing approaches find such attributions either using activations and gradients or by repeatedly perturbing the input. We instead address this challenge by training a second deep network, the Explainer, to predict attributions for a pre-trained black-box classifier, the Explanandum. These attributions are provided in the form of masks that only show the classifier-relevant parts of an image, masking out the rest. Our approach produces sharper and more boundaryprecise masks when compared to the saliency maps generated by other methods. Moreover, unlike most existing approaches, ours is capable of directly generating very distinct class-specific masks in a single forward pass. This makes the proposed method very efficient during inference. We show that our attributions are superior to established methods both visually and quantitatively with respect to the PASCAL VOC-2007 and Microsoft COCO-2014 datasets.