Uber researchers propose AI language model that emphasizes positive and polite responses

In a paper, Uber and Carnegie Mellon researchers propose a language model that can vary the positivity and politeness of its responses.