ImageBind

Google's open-weight EmbeddingGemma 2 model embeds text, code, images, video, and audio into a single shared vector space using a 270 million parameter Gemma-based text backbone plus optional 170M vis…

Researchers introduced LongAV-Compass, a benchmark for evaluating minute-long audio-visual generation across text, image, and video conditioning modalities. The benchmark contains 284 test cases and a…