Abstract: The visual classification performance of vision-language models such as CLIP has been shown to benefit from additional semantic knowledge from large language models (LLMs) such as GPT-3. In ...

AI outputs vary because confidence varies. Corroboration and entity optimization turn inconsistent AI visibility into consistent presence.

Abstract: We present a multi-way parallel corpus of Math Word Problems (MWPs) in nine languages, including six low-resource languages. To date, this is the largest multilingual MWP dataset available.

A math teacher recently became the top female Scrabble champion in the UK—scoring 1,000 points more than her opponents. Natalie Zolty outsmarted scrabble grandmaster Gary Oliver to claim the top spot ...