GPT is a Transformer-based architecture and training procedure for natural language processing tasks. Training follows a two-stage procedure. First, a language modeling objective is used on the unlabeled data to learn the initial parameters of a neural network model. Subsequently, these parameters are adapted to a target task using the corresponding supervised objective.
Source: Improving Language Understanding by Generative Pre-TrainingPaper | Code | Results | Date | Stars |
---|
Task | Papers | Share |
---|---|---|
Language Modelling | 84 | 10.78% |
Large Language Model | 50 | 6.42% |
Question Answering | 33 | 4.24% |
Prompt Engineering | 27 | 3.47% |
Text Generation | 23 | 2.95% |
Retrieval | 21 | 2.70% |
Sentence | 20 | 2.57% |
Decision Making | 19 | 2.44% |
In-Context Learning | 18 | 2.31% |