Models that combine text, images, audio and video.https://thelatentdaily.com/en-ushttps://thelatentdaily.com/research/vision-transformer-vit-explained/https://thelatentdaily.com/research/vision-transformer-vit-explained/In 2020, Google showed that a plain transformer applied to image patches can rival convolutional networks. How ViT works, why it needs lots of data and what it enabled.Sun, 18 Oct 2026 05:17:00 GMTComputer Visionresearchvision-transformercomputer-visiontransformereditors@thelatentdaily.comhttps://thelatentdaily.com/research/clip-contrastive-language-image-pretraining-explained/https://thelatentdaily.com/research/clip-contrastive-language-image-pretraining-explained/CLIP learned from 400 million image-caption pairs to match pictures with text, enabling zero-shot image classification and powering text-to-image generators.Sun, 18 Oct 2026 05:16:00 GMTMultimodal AIresearchclipmultimodalcomputer-visioneditors@thelatentdaily.comhttps://thelatentdaily.com/research/llava-visual-instruction-tuning-explained/https://thelatentdaily.com/research/llava-visual-instruction-tuning-explained/The 2023 LLaVA paper connected a vision encoder to an open language model and trained it on AI-generated visual instructions, creating an open multimodal assistant.Sun, 18 Oct 2026 05:15:00 GMTMultimodal AIresearchllavamultimodalopen-sourceeditors@thelatentdaily.comhttps://thelatentdaily.com/articles/vision-language-action-models-robotics/https://thelatentdaily.com/articles/vision-language-action-models-robotics/Robots that follow instructions like "pick up the extinct animal": how vision-language-action models such as RT-2 connect web knowledge to physical actions.Sun, 18 Oct 2026 04:33:00 GMTRobotics & Humanoidsroboticsvlaembodied-airesearcheditors@thelatentdaily.com