Introducing TabFM, a foundation model designed specifically for tabular data classification and regression. This approach allows generation of high-quality predictions on previously unseen tables in a single forward pass. Learn more and try out the model →https://goo.gle/3StxuLg
About us
From conducting fundamental research to influencing product development, our research teams have the opportunity to impact technology used by billions of people every day. We aspire to make discoveries that impact everyone, and sharing our research and tools to fuel progress in the field is fundamental to our approach.
- Website
-
https://research.google/
External link for Google Research
- Industry
- Technology, Information and Internet
- Company size
- 1,001-5,000 employees
Updates
-
Today we announce the expansion of our Heat Resilience data to cover 50+ global cities. By fusing satellite data, high-resolution imagery, and AI, we map building-level rooftop reflectivity to help urban planners mitigate extreme heat. Learn more: goo.gle/4b2605D
-
-
Today on the blog we introduce a method to retrofit Multi-Token Prediction onto frozen production models, accelerating on-device inference without the inefficiencies of separate drafters. Learn more →https://goo.gle/43XvmOo
-
-
Read all about how linear elastic caching minimizes total cache cost by framing page eviction as a ski rental problem, using lightweight machine learning to optimize the trade-off between memory footprint and cache misses. Check out the blog →https://goo.gle/3SoIXf3
-
-
Google researcher Dmitrii Kochkov was recently featured in Scientific American inaugural Young American Scientists list, as one of 28 early-career researchers whose work is shaping the future of science and society. He helped build the seasonal forecasting models that allowed Indian farmers to predict monsoons up to one month in advance — which could help ensure food security as climates change. #SciAmYoungScientists Learn more from this video: https://lnkd.in/gSwgJVYt Check out the article: https://lnkd.in/gaAiz4gb
He predicts the weather with machine learning - Dmitrii Kochkov - Young American Scientists 2026
https://www.youtube.com/
-
Today we present a study on how reasoning unlocks parametric knowledge in LLMs. We identify two key driving mechanisms, a computational buffer effect and factual priming, and suggest ways that can help build more reliable models. Learn more: goo.gle/4aioFdh
-
-
The world is experiencing a dramatic rise in extreme weather events and natural disasters. Over the past decade, Google has worked to make reliable information available to people at times of crises — when they need it most. 🌍 At yesterday’s AI for the Planet event, we shared the progress we’ve made toward building AI tools to support our vision for crisis resilience efforts: a world where no one is surprised by a natural disaster. By advancing AI research in flood forecasting, wildfire tracking, and extreme weather prediction, we are putting actionable intelligence directly into the hands of those who need it most. From our global flood models reaching 2 billion people to the development of FireSat for precise wildfire detection, see how we’re collaborating with partners to build a safer, more resilient future. Read more in a blog by Yossi Matias, GM of Google Research & Google Crisis Resilience Lead: https://goo.gle/3SJ928x
-
-
Evolving medical AI research from point-in-time diagnostics to long-term disease management requires a strong commitment to scientific rigor. Published today in Nature, our latest research with Google DeepMind introduces a new milestone for AMIE. In a randomized, blinded study using multi-visit scenarios, AMIE’s multi-agent architecture demonstrated longitudinal management reasoning on par with 21 primary care physicians. Read the full breakdown of how we're exploring the future of assistive medical AI below 👇
New publication from Google Research and Google DeepMind in Nature: We advance AMIE, our research medical AI, from one-off diagnostic conversations toward treating and managing disease over time, using clinical guidelines and drug formularies. In our randomized study, now published in Nature, we showed that AMIE demonstrated physician-level capabilities for treating and managing disease over sequential, multi-visit encounters with patient actors. The findings suggest that conversational medical AI systems like AMIE could one day help augment care and give doctors back time with their patients where it truly matters. While AMIE remains an experimental research system and is not ready for medical deployment, these peer-reviewed findings contribute to our strong commitment to evidence generation in AI for medical applications, offer an early look at how AI could help support long-term disease management. Learn more: https://goo.gle/3SaNXDW
-
Today we introduce a new vectorized dataset for mapping fine-scale ecological features, such as hedgerows, that often go undetected by standard satellites. This precision provides a new roadmap for addressing climate & biodiversity challenges without compromising food security. More: goo.gle/4fKKS7e
-
-
It is a golden age of research. "I have yet to find a problem that is too difficult to solve. It's a question of how much effort you need to put into that" - Yossi Matias, VP, Google & GM, Google Research. Some of our favorite highlights from the latest Release Notes include: → flood forecasting in 150 countries, covering 2B people for severe riverine floods, → the many applications of the MedGemma open model, → our geospatial models delivering critical weather warnings to 38 million farmers in India, → agentic systems like Co-Scientist reducing the timeline for generating a novel hypothesis from nearly a decade to just a couple of days, → and many other areas!
We’re living in a golden age of research ✨ I recently sat down with Logan Kilpatrick on the Release Notes podcast to dive into the incredible breadth of work happening across our teams at Google, and our overarching mission at Google Research of making the impossible possible. In this episode we discussed how AI serves as an amplifier of human ingenuity across so many critical areas: ✨ Accelerating science: using agentic tools like Co-Scientist, Empirical Research Assistance (ERA) and Paper Assistant Tool (PAT), to empower researchers and speed up scientific discovery. ✨ Transforming healthcare: How open platforms like MedGemma are already being used to provide access to healthcare solutions around the world. ✨ Planetary intelligence: Bringing our geospatial models together under Google Earth AI to transform planetary information into actionable intelligence. ✨ Education: reimagining text books with Learn Your Way, building models like LearnLM to adapt to learners goals and interests, aiming to empower teachers and bridge learning gaps for the next generation. We also discussed the magic cycle of research - from breakthrough research to transformative impact - in these as well as other areas, including generative AI - from efficiency and factuality to generative UI - and our progress in quantum computing. Hope you enjoy this discussion as much as I did: https://lnkd.in/eWCqZnfx