Common sense is not that common in AI, even with the researchers in the field of AI. “Right now, the recipe is, ‘Let’s just make the models super big – the bigger, the better,’ but humans – you and I – can’t really remember all that context, you know?” Yejin Choi is here to achieve the ‘mission impossible’ with AI, of giving summarising capabilities to AI models without RLHF, instead of just making them big.
Choi's research focuses on developing AI systems capable of commonsense reasoning, integrating visual and textual inputs to enhance language understanding. She has pioneered several key projects, including the creation of the ATOMIC knowledge graph and the COMET neural network system.
Her work has been recognised with numerous awards, including the MacArthur Fellowship, the Anita Borg Early Career Award, and multiple best paper awards at leading AI conferences.