VBART: The Turkish LLM

arXiv:2403.01308v1 Announce Type: cross
Abstract: We present VBART, the first Turkish sequence-to-sequence Large Language Models (LLMs) pre-trained on a large corpus from scratch. VBART are compact LLMs based on good ideas leveraged from BART and mBART models and come in two sizes, Large and XLarge. Fine-tuned VBART models surpass the prior state-of-the-art results in abstractive text summarization, title generation, text paraphrasing, question answering and question generation tasks. They allow fine-tuning for future text generation tasks and datasets, carving a new …

abstract art arxiv bart cs.ai cs.cl cs.lg good ideas language language models large language large language models llm llms prior results scratch state text type