H. Toprak Kesgin

A cyclical loss-based optimization algorithm for pretraining LLMs on noisy data

Knowledge-Based Systems

Turkish-e5: E5 Model Enhanced for Turkish with Multi-Positive Contrastive Learning

Creating a Large Clean Web Corpus for Turkish

Improving Information Retrieval in Turkish: A ColBERT-Based Approach

EAI/Springer Innovations in Communication and Computing

Optimal RAG System Design for Turkish Textbooks: A Comprehensive Evaluation and Performance Enhancement Study

Advancing NLP models with strategic text augmentation: A comprehensive study of augmentation methods and curriculum strategies

Natural Language Processing Journal

Introducing cosmosGPT: Monolingual Training for Turkish Language Models

Optimizing Large Language Models for Turkish: New Methodologies in Corpus Selection and Training

Cosmos-LLaVA: Chatting with the Visual

Scaling BERT Models for Turkish Automatic Punctuation and Capitalization Correction

Cyclical Curriculum Learning

IEEE Transactions on Neural Networks and Learning Systems

Iterative Mask Filling: An Effective Text Augmentation Method Using Masked Language Modeling

Communications in computer and information science

A robust optimization method for label noisy datasets based on adaptive threshold: Adaptive-k

Frontiers of Computer Science

Ensemble Learning Approach to Chatbot Design Based on Paraphrase Detection

Kocaeli Journal of Science and Engineering

Development and Comparison of Scoring Functions in Curriculum Learning

Investigating Semi-Supervised Learning Algorithms in Text Datasets

2022 Innovations in Intelligent Systems and Applications Conference (ASYU)