Sumários

Pre-trained models Applied

11 Novembro 2025, 11:00 Fernando Batista


- Encoder-based models: recent developments and trends

- Decoder-based models and LLMs: From GPT-1 to GPT-3, in-context learning

Pre-trained models Applied - Practical Examples

11 Novembro 2025, 09:30 Fernando Batista


Encoder-based models Applied: Adaptation of Pre-trained Encoder-based models to specific tasks
Training a GPT model from scratch: minGPT

Transformers

4 Novembro 2025, 11:00 Fernando Batista


Attention Mechanism, and Key designs (cont.)
Modifications and Improvements to the original transformer

transformers applied

4 Novembro 2025, 09:30 Fernando Batista


Pre-trained Models, Training objectives, downstream adaptations

Transformers

28 Outubro 2025, 11:00 Fernando Batista


Architecture Overview, Attention Mechanism, Key designs and innovations