Tantárgy adatlapja
Tárgy neve: Nagy nyelvmodellek létrehozása és alkalmazásai
Tárgy kódja: P_DO_0313
Óraszám: N: 2/2/0, L: 0/0/0
Kreditérték: 6
Az oktatás nyelve: angol
Követelmény típus: Kollokvium
Felelős kar: ITK
Felelős szervezeti egység: ITK Doktori és Habilitációs Iroda
Tárgyfelelős oktató:
Dr. Prószéky Gábor
Tárgyleírás:
Description of subject:
- Fundamentals of LLMs: understanding the architecture, training methods, and key components of LLMs
- Data preprocessing for LLMs: techniques and best practices for preparing datasets for training LLMs
- Model training and fine-tuning: in-depth exploration of training LLMs and fine-tuning them for specific applications
- Neural network architectures: examining transformer models and other neural network architectures used in LLMs
- Performance optimization: strategies for optimizing LLM performance, including hardware considerations and software optimizations
- Natural language understanding and generation: techniques for building applications that understand and generate human language
- Integration of LLMs in applications: best practices for integrating LLMs into real-world applications and systems
- Language model evaluation metrics: methods for evaluating the performance and accuracy of LLMs
- Scalable deployment of LLMs: approaches to deploying LLMs at scale, including cloud-based and edge deployment strategies
- Security and privacy in LLMs: ensuring data security and user privacy when developing and deploying LLM-based applications
- Future trends and innovations in LLMs: exploring cutting-edge research, emerging trends, and the future direction of LLM technology
List of required and recommended selected literature:
- Shervin Minaee, Tomas Mikolov, Narjes Nikzad, Meysam Chenaghlu, Richard Socher, Xavier Amatriain, Jianfeng Gao (2024) Large Language Models: A Survey. arXiv:2401.14423
- Shengyu Zhang, Linfeng Dong, Xiaoya Li, Sen Zhang, Xiaofei Sun, ShuheWang, Jiwei Li, Runyi Hu, Tianwei Zhang, Fei Wu and GuoyinWang (2024) Instruction Tuning for Large Language Models: A Survey. arXiv:2308.10792v8
List of required and recommended competences:
technical knowledge and skills, analytical and problem-solving skills, ethical and responsible ai development, integration and deployment skills, communication and collaboration, innovation and creativity
