Did you know that you can navigate the posts by swiping left and right?
With the development of modern LLMs (i.e. Large Languege Models), we've reached the turning point that the carefully trained model is more knowledgeable than most human-beings to some extent. The emergance of ChatGPT and recently published GPT4.0 sheds light on building Artificial General Intelligence in a feasible way In this blog, I would like to briefly introduce GPT-series and discuss its current limitations and future applications.
This is the first work to introduce Pre-Training - Finetuning paradigm into NLP domain.