Large Language Models like GPT-3 should be able to compete with specialized models on a lot of Natural Language Processing tasks without fine tuning ( Zero Shot Learning ) .
So I did a benchmark in order to verify that on four tasks : Keywords Extraction, Sentiment Analysis, language detection and translation against state of the art proprietary models from different companies like Google, Amazon, Microsoft, DeepL ...etc
Of course, with fine tuning the results should be better, but that's not the challenge here .
I'd love to know if anyone did this for other tasks like summarization or question answering ?
If you're tired of reading tons articles on how GPT3 is going to take over developers jobs. Here is one on how a dev can actually use it to make his app smarter (with code snippets).
