Researchers from the Chinese Academy of Science, including Q Fang, S Guo, Y Zhou, and Z Ma, have introduced LLaMA-Omni, a seamless speech interaction model using large language models. Another development in the field is LLaFS, a method for few-shot segmentation in computer vision, leveraging large language models. Additionally, GLaMM, a new AI model, can understand both text and images, enabling detailed conversations and object identification in pictures.