@google-research-blog
Publisher
102
Posts
A large proportion of clinical diagnoses can be derived from language-based interviews alone. These diagnostic interviews are typically conducted by clinicians through doctor-patient interactions during in-person or remote visits. While these interactions are the gold standard for symptom assessm...
Imagine a symphony orchestra performing a complex masterpiece. If the violins drifted out of tune every few measures, the ensemble would constantly have to stop and retune their instruments. Thankfully, this doesn't happen in an orchestra because the instruments reliably stay in tune. However, it...
Diffusion models are currently one of the most powerful types of tools for generative tasks that require complex and local structures, such as image generation and molecular discovery. They’ve shown an exciting capability to generalize beyond their training data and, in this sense, exhibit “creat...
Estimates suggest that billions of wearable devices are now in use, precisely tracking heart rate, movement, skin temperature, blood-oxygen levels, and sleep across days, weeks, and months. This continuous, longitudinal stream of physiology and behavior provides one of the most promising raw mate...
Vehicle transportation underpins much of modern life, enabling the movement of goods and people, productivity, and economic growth. However, the costs are high: drivers spend an average of 2.6 years of their life on the road, and private cars and vans now account for around 10% of global CO2 emis...
We’re releasing an expanded dataset of building-level rooftop reflectivity covering 50+ global cities to help urban planners implement cool-roof solutions and mitigate extreme heat. This dataset is being made accessible through a new high-resolution Heat Resilience Earth Engine App.
We’ve seen a massive shift in how people handle time-series forecasting since we launched TimesFM. Now, we’re bringing that same "zero-shot" logic to tabular data.
Having powerful Large Language Models (LLMs) right in your pocket is now a reality with on-device models like Gemini Nano and Gemma. This technology enables everyday features on your phone — such as instantly summarizing a flurry of notifications or proofreading an important text message — all wi...
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.
It is well-established that allowing large language models (LLMs) to generate step-by-step reasoning traces, commonly known as chain-of-thought (CoT), enhances performance on complex tasks. When a model solves difficult math equations, writes software, or answers multi-hop factual questions, brea...
We developed a high-resolution deep learning framework to reveal fine-scale ecological features, like hedgerows and copses, that are typically invisible to standard satellite detection. This precise vector data offers a new pathway to address the climate and biodiversity crises on working lands w...
More than half of adults use the Internet for health information, and one-third turn to artificial intelligence (AI). However, access to information does not mean that it is easy to understand or correctly interpreted. In short, the human component of AI for health information remains important t...
The carbon footprint of computing is a key sustainability challenge. It is driven by two major sources: operational carbon reflects emissions from energy consumed during use, and embodied carbon encompasses emissions associated with hardware manufacturing. While operational carbon is often addres...
Machine unlearning allows AI systems to "forget" specific parts of their training data without the massive cost of retraining a model from scratch. This is essential for regulatory compliance (like GDPR’s "Right to be Forgotten"), AI safety, and model quality.
Current single-step retrieval-augmented generation (RAG) systems weren’t designed for the multi-source, multi-hop queries of modern business workflows. If, for example, the query is, "What are the specs of the server used in Project X?", the system might find documents about Project X, but those...
Heart rate (HR), one of the cardinal vital signs, is a dynamic indicator of physiological status, influenced by everything from activity, to stress, to acute and chronic illness. Further, resting heart rate (RHR) is a key biomarker of cardiovascular health and long-term health risk. A higher RHR...
We have open-sourced our hydrology model to enable National Meteorological and Hydrological Services to integrate advanced AI-based flood forecasting into their own workflows.
This year’s breakthroughs at Google I/O reflect a new bold agentic era. With models that are more powerful than ever and an agentic coding platform, we’re making Google products substantially more helpful for everyone while transforming how researchers tackle the most pressing scientific and soci...
By processing data locally, on-device AI can provide enhanced protection and timely alerts while keeping user information private. For example, Android uses a system called SafetyCore to provide privacy-preserving on-device features and common infrastructure to protect users from unwanted content...
One of AI’s greatest potential benefits to humanity is increasing the speed and scope of scientific discovery. Empirical Research Assistance (ERA), a Google-developed research tool that uses Gemini to write and optimize scientific code, addresses one of the most time-consuming parts of scientific...
A scientific breakthrough reaches its full potential only when it empowers others to replicate and expand upon findings, pushing the boundaries of science even further. At Google Research, we recognize that open-source software and open-access datasets are drivers of modern science. We believe th...
AI’s capabilities to advance scientific discovery are growing every week, with outcomes that promise not just to enable breakthrough discoveries but to transform how science is done. In September, we released a preprint introducing Empirical Research Assistance (ERA) to help scientists generate e...
Have you ever looked back at your camera roll and wished you had captured a scene slightly differently? Maybe you wish you had caught a bit more of one side of a face, or positioned the camera slightly lower to get the perfect shot. Perhaps it’s a selfie with a perfect smile, but the wide-angle l...
Agents are becoming increasingly crucial in tackling complex real-world tasks, ranging from general web navigation to assisting with extensive software engineering codebases. However, as these agents transition into persistent, long-running roles in the real world, they face a critical limitation...
The rapid advance of generalist AI models has been fueled by the abundance of internet data. However, widespread integration of AI will require models to specialize in novel, uncommon, and privacy-sensitive applications where data is inherently scarce or inaccessible.
Using computers to create full wiring maps of complex brains is enabling a new era of neuroscience. A recently released map of the complete male fruit fly brain and central nervous system provides a foundational resource for studying how the brain responds to stimuli and controls the body.
As AI evolves at an unprecedented pace, there is a renewed focus on "future-ready" skills — the durable human competencies that will remain valuable regardless of technological shifts or automation. International frameworks, such as the OECD Learning Compass 2030 and the WEF’s Future of jobs repo...
Modern conversational AI agents can typically handle complex, multi-turn tasks like asking clarifying questions and proactively assisting users. However, they frequently struggle with long interactions, often forgetting constraints or generating irrelevant responses. Improving these systems requi...
Academic research is evolving at an unprecedented pace driven by the rapid advancements in AI. The academic research workflow is notoriously rigorous, involving far more than just conceptualizing an idea and writing a paper. One hurdle many researchers face is how to effectively visualize their r...
As LLMs integrate into our daily lives, understanding their behavior becomes essential. In our ongoing efforts to study model behavior and alignment, we present this work as an early step in that direction. We focus on behavioral dispositions — the underlying tendencies that shape responses in so...
Google Research explores the trade-off between number of items and human raters per item to improve AI benchmark reproducibility and capture the nuance of human disagreement.
Large language models (LLMs) and agentic workflows are changing software engineering and creative computing. We are seeing a shift toward “vibe coding”, where LLMs turn human intent directly into working code. Tools like Gemini Canvas already make this possible for 2D and 3D web development. Howe...
Vectors are the fundamental way AI models understand and process information. Small vectors describe simple attributes, such as a point in a graph, while “high-dimensional” vectors capture complex information such as the features of an image, the meaning of a word, or the properties of a dataset....
When we think about artificial intelligence and geography, we often focus on navigation, or getting from point A to point B. However, the built environment — the complex web of roads, buildings, businesses, and infrastructure that defines our world — contains far more information than just coordi...
For a decade, we’ve been advancing fundamental computer science research to address real-world healthcare challenges, in pursuit of scientific and clinical discovery. While the pace of AI research today is staggering, our sense of responsibility remains steadfast. We uphold the highest standards...
Breast cancer is the leading cause of death for women aged 35–64 in the UK, but research has conclusively shown that early screening through mammography saves lives. The UK's National Health Service (NHS) Breast Screening Programme currently relies on a double-read workflow — two human mammograph...
Artificial intelligence (AI) is now routinely used to compose emails, edit images and summarize information from the web. AI also holds enormous potential to accelerate scientific research. However, its effectiveness in providing scientifically accurate and comprehensive answers to complex questi...
According to the World Meteorological Organization (WMO), flash floods account for approximately 85% of flood related fatalities worldwide. They typically occur within six hours of heavy rain, turn city streets into gushing rivers, and take more than 5,000 lives annually, making them one of the w...
Today, we’re introducing Groundsource, a new scalable methodology that leverages Gemini to transform unstructured global news into actionable, historical data. Our first, open-access Groundsource dataset for urban flash floods comprises 2.6 million records, paving the way for more accurate, life-...
AI systems capable of clinical reasoning and dialogue have the potential to dramatically increase access to medical expertise and care while giving physicians back time with their patients. Yet, bringing these innovations to life demands a safety-centric, evidence-based approach. In recent years,...
WAXAL provides a critical, open-access foundation for African speech technology. Featuring a large corpus of ASR and TTS data for 27 native languages under a highly permissive license, WAXAL empowers the African AI ecosystem to build robust speech systems that better reflect the region's unique l...
One year after going open-source, Google’s SpeciesNet model is accelerating wildlife conservation by identifying nearly 2,500 species in camera trap images globally. Learn how this AI tool supports biodiversity research from the global research community.
Google researchers demonstrate how Bayesian teaching through supervised fine-tuning enables LLMs to approximate optimal probabilistic reasoning and generalize to new domains.
Look at a map of a shopping mall or a theme park. Within seconds, your brain processes the visual information, identifies your location, and traces the optimal path to your destination. You instinctively understand which lines are walls and which are walkways. This fundamental skill — fine-graine...
In the world of algorithmic job scheduling, computing resources are often viewed as static: a server has a fixed number of CPUs, or a cluster has a constant number of available machines. However, the reality of modern large-scale cloud computing is far more dynamic. Resources fluctuate constantly...
Conversational AI has fundamentally reshaped how we interact with technology. While one-on-one interactions with large language models (LLMs) have seen significant advances, they rarely capture the full complexity of human communication. Many real-world dialogues, including team meetings, family...
Discover how Google researchers repurposed AI models trained on bird songs to detect whale calls and track marine health. Explore the future of bioacoustic monitoring.
At Google, we believe in building for everyone and accessibility (A11y) is a key part of that. Our teams work with communities to build products with and for people with disabilities, incorporating accessibility from the beginning of the development process. Today, generative AI provides us with...
Feature selection is the process of identifying and retaining the most informative subset of input variables while discarding irrelevant or redundant noise. A fundamental challenge in both machine learning and deep learning, feature selection is NP-hard (i.e., a problem that is mathematically "im...
In partnership with Included Health, we will be launching a first-of-its-kind nationwide study to evaluate conversational AI within real-world virtual care workflows. This research will move beyond simulation and retrospective data and aim to gather rigorous prospective evidence on how AI perform...
AI agents — systems capable of reasoning, planning, and acting — are becoming a common paradigm for real-world AI applications. From coding assistants to personal health coaches, the industry is shifting from single-shot question answering to sustained, multi-step interactions. While researchers...
Google Research introduces ATLAS, a new framework for predicting optimal model size and data mixtures across 400+ languages to solve the "curse of multilinguality".
Google researchers introduce GIST, a novel algorithm that balances data diversity and utility with provable guarantees for efficient ML model training.
As AI technologies advance, truly helpful agents will become capable of better anticipating user needs. For experiences on mobile devices to be truly helpful, the underlying models need to understand what the user is doing (or trying to do) when users interact with them. Once current and previous...
Gait metrics — measures like walking speed, step length, and double support time (i.e., the proportion of gait cycle when both feet are on the ground) — are known to be vital biomarkers for assessing a person’s overall health, risk of falling, and progression of neurological or musculoskeletal co...
We establish a positive association between hard-braking events (HBEs) collected via Android Auto and actual road segment crash rates. We confirm that roads with a higher rate of HBEs have a significantly higher crash risk and suggest that such events could be used as leading measures for road sa...
The adoption of artificial intelligence in healthcare is accelerating dramatically, with the healthcare industry adopting AI at twice the rate of the broader economy. In support of this transformation, last year Google published the MedGemma collection of open medical generative AI models through...
Quantum error correction (QEC) is crucial for reaching the ultra low error rate necessary for useful quantum algorithms. At Google Quantum AI, our quantum processors use physical qubits constructed from small superconducting circuits, which are susceptible to noise. QEC allows us to combine numer...
Precipitation remains one of the trickiest tasks for global-scale weather and climate models. That’s because exactly where, when and how much precipitation will fall depends on a series of events happening at scales that are typically below the model resolution. Simulating precipitation is especi...
Google Research teams have invested over the years in advancing research and technology in a diverse range of strategic areas. We are working across time horizons, from bold moonshots and curiosity-driven transformative research where we explore the art of the possible, to innovation and applied...
The pursuit of truth in theoretical computer science and mathematics relies on the highest standards of proof, rigor, and clarity. While peer review is the crucial final check, the process of drafting and refining complex theoretical work often takes months, with simple errors, inconsistent varia...
Generative AI is rapidly changing healthcare, opening up opportunities to better address real-world health challenges. Across the African continent, there is broad interest in tackling such challenges, ranging from cervical cancer screening to maternal health support.
Large language model (LLM) chatbots are used by hundreds of millions of people daily for tasks ranging from drafting emails and writing code to planning vacations and creating menus for cafes. Understanding these high-level use cases is incredibly valuable for platform providers looking to improv...
The Transformer architecture revolutionized sequence modeling with its introduction of attention, a mechanism by which models look back at earlier inputs to prioritize relevant input data. However, computational cost increases drastically with sequence length, which limits the ability to scale Tr...
Sound is a critical part of multimodal perception. For a system — be it a voice assistant, a next-generation security monitor, or an autonomous agent — to behave naturally, it must demonstrate a full spectrum of auditory capabilities. These capabilities include transcription, classification, retr...
The transition to electric vehicles (EVs) is accelerating globally, bringing with it the critical need for a reliable and robust charging infrastructure. While building out more physical charging stations is an important step, an equally important task is maximizing the efficiency of this infrast...
Real-time communication is an integral part of both our professional and personal lives. When speaking to people remotely across language barriers, it can be difficult to truly connect by just relying on state-of-the-art translated captions, as they lack personality and real-time responsiveness e...
Generative UI is a powerful capability in which an AI model generates not only content but an entire user experience. Today we introduce a novel implementation of generative UI, which dynamically creates immersive visual experiences and interactive interfaces — such as web pages, games, tools, an...
Forests are vital for our planet as they regulate rainfall, mitigate floods, store and sequester carbon, and help sustain the majority of the planet’s land-based species. Despite their importance, deforestation continues at an alarming rate. A key challenge in conservation efforts is differentiat...
From designing more efficient airline routes to organizing clinical trials, optimization problems are everywhere. Yet for many real-world challenges, even our most powerful supercomputers struggle to find the best solution. This has led to a major, decades-long question in quantum computing: coul...
From personalized recommendations to scientific advances, AI models are helping to improve lives and transform industries. But the impact and accuracy of these AI models is often determined by the quality of data they use. Large, high-quality datasets are crucial for developing accurate and repre...
The last decade has seen incredible progress in machine learning (ML), primarily driven by powerful neural network architectures and the algorithms used to train them. However, despite the success of large language models (LLMs), a few fundamental challenges persist, especially around continual l...
Data science is a field dedicated to transforming raw data into meaningful, actionable insights, playing an essential role in solving real-world challenges. Businesses often depend on data-driven insights to make pivotal strategic decisions. However, the data science process is frequently complex...
Nature underpins our climate, our economies, and our very lives. And within nature, forests stand as one of the most powerful pillars — storing carbon, regulating rainfall, mitigating floods, and harboring the majority of the planet’s terrestrial biodiversity.
Artificial intelligence (AI) is a foundational technology that could reshape our world, driving new scientific discoveries and helping us tackle humanity's greatest challenges. Now, we're asking where we can go to unlock its fullest potential.
Last week at our flagship Research@ event in Mountain View, we shared some of Google Research’s latest announcements, from understanding earth to advancements in genomics to advancements in quantum computing. Working collaboratively with colleagues across the company, our teams drive breakthrough...
Generative AI (GenAI) enables personalized experiences and powers the creation of unstructured data, including summaries, transcriptions, and more. Insights into real-world AI use [1, 2] can help GenAI developers enhance their tools by understanding common applications and identifying failure mod...
Interactive streetscape tools, available today in every major mapping service, have revolutionized how people virtually navigate and explore the world — from previewing routes and inspecting destinations to remotely visiting world-class tourist locations. But to date, screen readers have not been...
The personal health coach is built with Gemini models to deliver personalized and adaptive coaching, grounded in science and informed by expert oversight.
For years, Google has developed AI models that enhance our understanding of the planet. These models help keep Google products fresh, for example, ensuring Maps is accurate by analyzing satellite images and giving Search users the most up-to-date alerts about weather and natural disasters.
Nature is brimming with chaos, a phenomenon characterized by the high sensitivity of a system toward small perturbations. In the macroscopic world, notable examples of chaotic systems include weather patterns, wherein a small change in initial conditions leads to vastly different outcomes over ti...
Differential privacy (DP) provides a powerful, mathematically rigorous assurance that sensitive individual information in a dataset remains protected, even when a dataset is used for analysis. Since DP’s inception nearly two decades ago, researchers have developed differentially private versions...
Modern astronomy is a treasure hunt on a cosmic scale. Every night, telescopes around the globe scan the skies, searching for fleeting events like exploding stars (supernovae) that give us crucial insights into the workings of the universe. These surveys generate millions of alerts about potentia...
Imagine a puzzle game similar to Tetris with pieces rapidly falling onto a stack. Some fit perfectly. Others don’t. The goal is to pack the blocks as tightly and efficiently as possible. This game is a loose analogy to the challenge faced by cloud data centers several times every second as they t...
Cancer is fundamentally a genetic disease in which the genetic controls on cell division go awry. Many types of cancer exist, and each poses unique challenges as it can have distinct genetic underpinnings. A powerful way to study cancer, and a critical step toward creating a treatment plan, is to...
Generative AI has fundamentally reshaped our expectations of technology. We've seen the power of large-scale cloud-based models to create, reason and assist in incredible ways. However, the next great technological leap isn't just about making cloud models bigger; it's about embedding their intel...
The combination of artificial intelligence (AI) and extended reality (XR) has the potential to unlock a new paradigm of immersive intelligent computing. However, a significant gap exists between the ecosystems of these two fields today. AI research and development is accelerated by mature framewo...
Voice-based web search has been around a long time and continues to be used by many people, with the underlying technology evolving rapidly to allow for expanded use cases. Google’s initial voice search solution used automatic speech recognition (ASR) to turn the voice input into a text query, an...
You have a perfect image in your mind. You enter a prompt, hit generate, and the result is close to what you were thinking, but not quite right. You try refining the prompt, adding more detail, but you can't seem to bridge the gap between your idea and the final image.
The key to elevating a good photo often lies in selective image adjustments: brightening a subject in the foreground, enhancing the sky, or making the color of a jacket pop. Yet, isolating specific elements with existing tools that offer subject, background, sky, or color-based selections has rem...
Recently, large language models (LLMs) have demonstrated surprising capabilities in competitive mathematics and competitive programming, demonstrating world-leading performance across both of these fields. However, their successes in mathematical discovery — proving novel theorems or uncovering n...
The rapid advancement of large language models (LLMs), combined with data from wearable devices, presents a transformative opportunity to empower people on their personal health journeys. However, health needs vary from individual to individual. Answering a specific query, such as, "On average, h...
The ability to find clear, relevant, and personalized health information is a cornerstone of empowerment for medical patients. Yet, navigating the world of online health information is often a confusing, overwhelming, and impersonal experience. We are met with a flood of generic information that...
Large language models (LLMs) have shown potential for medical and health question answering across various health-related tests spanning different formats and sources, such as multiple choice and short answer exam questions (e.g., USMLE MedQA), summarization, and clinical note taking, among other...
Time-series forecasting is essential for modern businesses, helping them predict everything from inventory needs to energy demands. Traditionally, this has involved building a separate, specialized model for each task — a process that is slow and requires significant expertise.
The recent advances in large language models (LLMs) have fueled the emergence of deep research (DR) agents. These agents demonstrate remarkable capabilities, including the generation of novel ideas, efficient information retrieval, experimental execution, and the subsequent drafting of comprehens...
Recent innovations, such as Google's Project Astra, exemplify the potential of proactive agents embedded in augmented reality (AR) glasses to offer intelligent assistance that anticipates user needs and seamlessly integrates into everyday life. These agents promise remarkable convenience, from ef...
Large language models (LLMs) have come a long way and achieved some remarkable breakthroughs in recent years. However, they sometimes have issues with factuality, confidently making claims that are incorrect. Known as “hallucination”, this issue arises from a number of factors, including incomple...
Textbooks are a cornerstone of education, but they have a fundamental limitation: they are a one-size-fits-all medium. The manual creation of textbooks demands significant human effort, and as a result they lack alternative perspectives, multiple formats and tailored variations that can make lear...
As AI becomes more integrated into our lives, building it with privacy at its core is a critical frontier for the field. Differential privacy (DP) offers a mathematically sound solution by adding calibrated noise to prevent memorization. However, applying DP to LLMs introduces trade-offs. Underst...
LLMs have transformed how we interact with technology, powering everything from advanced search capabilities to creative coding assistants. But this power comes at a cost: inference (the process of generating a response) can be slow and computationally expensive. As we deploy these models to more...
