Amazon has unveiled a new language model that improves GPT-3.5’s performance on ScienceQA by a margin of 16 percentage points, or from 75.1% accuracy to 91.6% accuracy. Will Amazon lead the way in the chatbot competition and become a participant? (Representational image: Tobias Dziuba via Pexels)
Read Also: ChatGPT Vs Google Bard
A little over two months ago, OpenAI presented ChatGPT to the general public for the first time. Almost immediately, the AI-powered chatbot was thrust into the center of mainstream discourse, with discussions revolving around the ways in which it could impact business, education, and other fields.
Then, internet giants Google and Baidu, based in China, debuted their chatbots to convince the general public that their so-called “generative AI” (technology that can produce conversational text, images, and more) is also ready for prime time.
Read Also: ChatGPT Chrome Extensions
At this stage, on the ScienceQA benchmark, Amazon’s new language models perform better than GPT-3.5 by 16 percentage points (75.17%) than GPT-3.5, and they even outperform many humans.