We introduce JEEM, a benchmark designed to evaluate Vision-Language Models (VLMs) on visual understa
This paper presents an Amharic hate speech annotation using the Yandex Toloka crowdsourcing platform