VideoPrism: A foundational visual encoder for video understanding
Test Information Space
SCIN: A new resource for representative dermatology images
Cappy: Outperforming and boosting large multi-task language models with a small scorer
Gemini can now access and create Google Calendar events
(The BuJo, SloMo, or RoBo methods seem to start small to see the whole picture and then scope in. That may look like a single page showing the week, month, quarter, or year's dates to get the values/goals/priorities then the master task list and then distributing them across the two-page spreads down to day times. People used to do this shuffling Rolodex cards into order, e.g. next contact rather than action. You can use a whiteboard, too. Or the planners also have dedicated book formats. Lists get their own pages by topic, e.g. reading, shopping, or cooking. Besides taxes, businesses also schedule social media. For sanity, the handwritten and digital would match formats such as scheduling columns. Books can be rotated sideways. Google Calendar has multiple views and integrates with other apps. MS Copilot finds meeting times.)
Health-specific embedding tools for dermatology and pathology
Croissant: a metadata format for ML-ready datasets
Google brings Stack Overflow's knowledge base to Gemini for Google Cloud | TechCrunch
Genie: Generative Interactive Environments
Announcing Gemini for Google Workspace | Google Workspace Blog
Advances in private training for production on-device language models
Building Open Models Responsibly in the Gemini Era
Google Is Giving Away Some of the A.I. That Powers Chatbots
Deepfake fears intensify as Google CEO sounds alarm on AI misinformation
Google’s AI Boss Says Scale Only Gets You So Far
Our next-generation model: Gemini 1.5
Gemini - Google DeepMind
Google Gemini Ultra 1.0 Complete Review - Worth the Upgrade?
The next chapter of our Gemini era
Graph neural networks in TensorFlow
A decoder-only foundation model for time-series forecasting
Bard’s latest updates: Access Gemini Pro globally and generate images
Google's AI Makes Stunning Progress with Logical Reasoning
DSI++: Updating Transformer Memory with New Documents
Lumiere: A Space-Time Diffusion Model for Video Generation
Chrome is getting 3 new generative AI features
Steven Johnson on Twitter / X
Can large language models identify and correct their mistakes?
Google faces $1.67 bln damages demand at AI-related patent trial
Responsible AI at Google Research: User Experience Team
Here's what Google Assistant with Bard will look like