2024 · 51 citations · 19 references
Gemini models excel in various tasks including image generation and interpretation, video understanding, and solving mathematical problems, among others. The Vertex AI Gemini API and Google AI Gemini API both enable developers to integrate Gemini model functionalities into their applications. This paper offers a concise summary of the Gemini Framework, focusing on its distinctive modalities that distinguish it from current systems. In our research, we explored the details of its architecture, pointing out the innovative strategies employed to improve generative AI capabilities. Furthermore, we conduct a comparative study, assessing Gemini’s performance against other top generative AI models.
19
Robust Speech Recognition via Large-Scale Weak Supervision
Alec Radford, Jong Wook Kim, Tao Xu et al. · arXiv (Cornell University) · 2022 · 1.2K citations · Full text
ImageBind One Embedding Space to Bind Them All
Rohit Girdhar, Alaaeldin El-Nouby, Zhuang Liu et al. · 2023 · 712 citations
LaMDA: Language Models for Dialog Applications
Romal Thoppilan, Daniel De Freitas, Jamie Hall et al. · arXiv (Cornell University) · 2022 · 706 citations · Full text