In this paper, we consider the model merging process for large language models (LLMs) under a two-stage optimization framework. Traditional merging methods usually apply fixed blending rates to all layers, which ignores the different levels of contribution of layers to the output. To overcome this limitation, we first group layers with similar functionality and use Bayesian optimization to determine the optimal blending rates for each group. Bayesian optimization offers a globally effective discovery strategy in high-dimensional and computationally expensive search spaces. However, an approach based solely on global search can lead to local artifacts being missed. Therefore, in the second stage, we perform Linear Search by applying controlled variations around the ratios obtained with Bayesian optimization and obtain more precise, locally optimized results. Extensive experiments on the Turkish versions of the Arc, Hellaswag, MMLU and GSM8K datasets show that our proposed Bayes+Linear strategy outperforms existing methods such as SLERP, Linear, Ties and Breadcrumbs. In addition to improving model accuracy and generalization capacity, the KAFA approach works without any additional fine-tuning, making it possible to effectively reuse large language models in different tasks.

PurposeThe aim of this study was to evaluate the performance of the YOLOv11 deep learning–based object detection model in detecting dental restorations on digital panoramic radiographs.Materials and MethodsA total of 320 panoramic radiographs obtained from adult patients were included in this study. Four types of dental restorations—amalgam, composite, crown, and bridge—were manually annotated and used for model training and evaluation. The dataset was divided into training and testing subsets at a ratio of 75% and 25%, respectively. A validation subset was derived from the training data. Model performance was assessed using precision, recall, F1-score, and mean Average Precision (mAP) metrics.ResultsThe YOLOv11 model achieved an overall precision of 0.663, recall of 0.665, and F1-score of 0.663 across all restoration categories. The mAP@50 value was 0.671, while the mAP@50–95 value was 0.432, indicating variations in detection performance across different Intersection over Union thresholds. Differences in detection performance were observed among the various types of dental restorations.ConclusionThe findings indicate that YOLOv11 demonstrates measurable potential as an automated support tool for detecting dental restorations on panoramic radiographs; however, the achieved performance remains lower than that reported in several previous studies. Successful clinical integration of such AI-assisted systems requires further optimization and validation using larger and more heterogeneous datasets to enhance generalizability and clinical applicability.

In this study, we develop and assess new corpus selection and training methodologies to improve the effectiveness of Turkish language models. Specifically, we adapted Large Language Model generated datasets and translated English datasets into Turkish, integrating these resources into the training process. This approach led to substantial enhancements in model accuracy for both few-shot and zero-shot learning scenarios. Furthermore, the merging of these adapted models was found to markedly improve their performance. Human evaluative metrics, including task-specific performance assessments, further demonstrated that these adapted models possess a greater aptitude for comprehending the Turkish language and addressing logic-based queries. This research underscores the importance of refining corpus selection strategies to optimize the performance of multilingual models, particularly for under-resourced languages like Turkish.

This study introduces and evaluates tiny, mini, small, and medium-sized uncased Turkish BERT models, aiming to bridge the research gap in less-resourced languages. We trained these models on a diverse dataset encompassing over 75GB of text from multiple sources and tested them on several tasks, including mask prediction, sentiment analysis, news classification, and, zero-shot classification. Despite their smaller size, our models exhibited robust performance, including zero-shot task, while ensuring computational efficiency and faster execution times. Our findings provide valuable insights into the development and application of smaller language models, especially in the context of the Turkish language.

Knowledge distillation (KD) transfers knowledge from large language models (LLMs) to smaller or similarly sized models in order to obtain efficient yet capable systems. However, performing distillation over all tokens is computationally expensive and may weaken the transfer signal. To address this limitation, Knowledge-Filtered Distillation (KFD) is introduced as a selective distillation approach in which tokens are filtered according to the divergence $KL({\mathcal{M}}_2\mid {\mathcal{M}}_0)$ between a teacher model (${\mathcal{M}}_2$) and a base model (${\mathcal{M}}_0$), while the student model (${\mathcal{M}}_1$) is also derived from the same base model. Only tokens whose divergence exceeds a predefined threshold are distilled. For the selected tokens, the teacher distribution is normalized over the Top-5 predictions, whereas tokens outside this case receive a label-ranking bonus. The proposed conditional Top-5/bonus target design is shown theoretically to yield a lower label-focused target error than using only Top-5 normalization or only the bonus across all tokens. In addition, the KL and cross-entropy (CE) losses are balanced through a dynamically computed batch-level coefficient $\alpha$. Experiments on multiple Turkish text datasets show that KFD consistently outperforms CE-only training, achieving higher accuracy with less data and shorter training time. KFD also outperforms entropy-based token selection methods and highlights the role of student initialization in effective knowledge transfer, thereby providing an efficient and scalable distillation framework for teacher–student models of equal size.