Current research in yes/no question answering (QA) focuses on transfer learning techniques and transformer-based models. Models trained on large corpora are fine-tuned on tasks similar to yes/no QA, and then the captured knowledge is transferred for solving the yes/no QA task. Most previous studies use existing similar tasks, such as natural language inference or extractive QA, for the fine-tuning step. This paper follows a different perspective, hypothesizing that an artificial yes/no task can transfer useful knowledge for improving the performance of yes/no QA. We introduce three such tasks for this purpose, by adapting three corresponding existing tasks: candidate answer validation, sentiment classification, and lexical simplification. Furthermore, we experimented with three different variations of the BERT model (BERT base, RoBERTa, and ALBERT). The results show that our hypothesis holds true for all artificial tasks, despite the small size of the corresponding datasets that are used for the fine-tuning process, the differences between these tasks, the decisions that we made to adapt the original ones, and the tasks’ simplicity. This gives an alternative perspective on how to deal with the yes/no QA problem, that is more creative, and at the same time more flexible, as it can exploit multiple other existing tasks and corresponding datasets to improve yes/no QA models.

This study aimed to investigate the incidence of subclinical mastitis (SCM), the implicated pathogens, and their impact on milk quality in dairy sheep in Greece. Furthermore, we preliminarily evaluated infrared thermography and the application of AI tools for the early, non-invasive diagnosis of relevant cases. In total, 660 milk samples and over 2000 infrared thermography images were obtained from 330 phenotypically healthy ewes. Microbiological investigations, a somatic cell count (SCC), and milk chemical analyses were performed. Infrared images were analyzed using the FLIR Research Studio software (version 3.0.1). The You Only Look Once version 8 (YOLOv8) algorithm was employed for the automatic detection of the udder’s region of interest. A total of 157 mammary glands with SCM were identified in 122/330 ewes (37.0%). The most prevalent pathogen was staphylococci (136/160, 86.6%). Considerable resistance was detected to tetracycline (29.7%), ampicillin (28.6%), and sulfamethoxazole–trimethoprim (23.6%). SCM correlated with high total mesophilic count (TMC) values and decreased milk fat, lactose, and protein content. A statistically significant variation (p < 0.001) was identified in the unilateral SCM cases by evaluating the mean temperatures of the udder region between the teats in the thermal images. Finally, the YOLOv8 algorithm was employed for the automatic detection of the udder’s region of interest (ROI), achieving 84% accuracy in defining the ROI in this preliminary evaluation. This demonstrates the potential of infrared thermography combined with AI tools for the diagnosis of ovine SCM. Nonetheless, more extensive sampling is essential to optimize this diagnostic approach.

The effectiveness of natural language processing models relies on various factors, including the architecture, number of parameters, data used during training, and the tasks they were trained on. Recent studies indicate that models pre-trained on large corpora and fine-tuned on task-specific datasets, covering multiple tasks, can generate remarkable results across various benchmarks. We propose a new approach based on a straightforward hypothesis: improving model performance on a target task by considering other artificial tasks defined on the same training dataset. By doing so, the model can gain further insights into the training dataset and attain a greater understanding, improving efficiency on the target task. This approach differs from others that consider multiple pre-existing tasks on different datasets. We validate this hypothesis by focusing on the problem of answering yes/no questions and introducing a multi-task model that outputs a span of the reference text, serving as evidence for answering the question. The task of span extraction is an artificial one, designed to benefit the performance of the model answering yes/no questions. We acquire weak supervision for these spans, by using a pre-trained extractive question answering model, dispensing the need for costly human annotation. Our experiments, using modern transformer-based language models, demonstrate that this method outperforms the standard approach of training models to answer yes/no questions. Although the primary objective was to enhance the performance of the model in answering yes/no questions, it was discovered that span texts are a significant source of information. These spans, derived from the question reference texts, provided valuable insights for the users to better comprehend the answers to the questions. The model’s improved accuracy in answering yes/no questions, coupled with the supplementary information provided by the span texts, led to a more comprehensive and informative user experience.