Pinterest Labs is the home of our ML research and AI product innovation
Across recommender systems, visual foundation models, graph understanding, and responsible AI research, Pinterest Labs has been investing in ML technology for more than a decade.
Recommender systems
TransAct is the ranking system that powers Pinterest’s home feed, taking into account both real-time engagement data and long user activity sequences via a hybrid architecture.
Generative models
Pinterest Canvas is a text-to-image diffusion model trained on the visual graph data of Pinterest, allowing for the ability for merchants to create personalized lifestyle photography out of their product catalog.
Representation learning
OmniSearchSage is a unified embedding system for representing Pin items, search queries, and products.
Machine Learning powers our mission
Pinterest’s mission is to provide everyone with the inspiration to create a life they love, and Machine Learning sits at the core of making that mission a reality.
Machine Learning is not only powering predictive results for people across interests to help them find their next great idea, it’s identifying misinformation, surfacing relevant ads, and advancing inclusivity and body positivity, to name a few areas. Pinterest engineers are building with Machine Learning to make Pinterest representative of the people who use it every day, forging the way for a positive corner of the internet we can all be proud of, and find useful.
Watch the video or view the audio described video here.
Other innovations at the lab
Shopping / Creators
Shopping on Pinterest feels a lot like selling products in the real world. Share a gorgeous catalog, set up a storefront, and then get to the good stuff, like tracking conversions and measuring results.
Inclusive search
Powering inclusive search & recommendations with our new visual skin tone model.
Pinterest Lens
Shop with your camera: Pinterest launches Shop tab on Lens visual search results