The last couple of years have seen a very significant improvement in the performance of “Large Language Models” like GPT-3 and now ChatGPT
Yes, it is true. It is impressive. Since the “Attention is all you need” paper published in 2017, there’s been a focus on Transformers that have really changed the landscape of AI in many fronts. I think the authors of the Attention paper never imagined that their architecture would cause such a revolutionary impact in the field. And I think that this is just the begining.
This is a quote from this other recent thread, which is also worth a look.
Hey @Juan_Olano, do you mind posting a link to the paper? I’d love to take a look at it. I’m sure other community members would love to as well.
Hi @deltannboi ,
For a great introduction to this model I recommend considering the Deep Learning Specialization. In its 5th course, Sequential models, prof. Ng explains very clearly the Attention model.
True. I think it will improve significantly in the future to even make large reports on critical events.