Publications
Our teams aspire to make discoveries that impact everyone, and core to our approach is sharing our research and tools to fuel progress in the field.
Our teams aspire to make discoveries that impact everyone, and core to our approach is sharing our research and tools to fuel progress in the field.
Sort By
1 - 15 of 10793 publications
Preview abstract
For many practical applications of quantum computing, the slowest and most costly steps involve coherently accessing classical data. We help address this challenge by applying mass production techniques, which can sometimes allow us to perform operations many times in parallel for a cost that is comparable to a single execution[1-3]. We combine existing mass-production results with modern approaches for loading classical data using ``quantum read-only memory.'' We show that quantum mass production techniques offer no benefit when we consider a cost model that focuses purely on the number of non-Clifford gates. However, analyzing the constant factors in a more nuanced cost model, we find that it may be possible to obtain a reduction in cost of an order or magnitude or more for a variety reasonably-sized fault-tolerant quantum algorithms. We present several applications of quantum mass-production techniques beyond naive parallelization, including a strategy for reducing the cost of serial calls to the same data loading step.
View details
Preview abstract
AI coding assistants are rapidly becoming integral to modern software development. A key challenge in this space is the continual need to migrate and modernize codebases in response to evolving software ecosystems. Traditionally, such migrations have relied on rule-based systems and human intervention. With the advent of powerful large language models (LLMs), AI-driven agentic frameworks offer a promising alternative—but their effectiveness remains underexplored. In this paper, we introduce FreshBrew, a novel benchmark for evaluating AI-based agentic frameworks on project-level Java migrations. We benchmark several such frameworks, powered by state-of-the-art LLMs, and compare their performance against established rule-based tools. Our evaluation of AI agents on this benchmark of 228 repositories shows that the top-performing model, Gemini 2.5 Flash, can successfully migrate 56.5% of projects to JDK 17. Our empirical analysis reveals novel insights into the critical strengths and limitations of current agentic approaches, offering actionable insights into their real-world applicability. By releasing FreshBrew publicly upon acceptance, we aim to facilitate rigorous, reproducible evaluation and catalyze progress in AI-driven codebase modernization.
View details
Preview abstract
Understanding fine-grained temporal dynamics is crucial in egocentric videos, where continuous streams capture frequent, close-up interactions with objects. In this work, we bring to light that current egocentric video question-answering datasets often include questions that can be answered using only few frames or commonsense reasoning, without being necessarily grounded in the actual video. Our analysis shows that state-of-the-art Multi-Modal Large Language Models (MLLMs) on these benchmarks achieve remarkably high performance using just text or a single frame as input. To address these limitations, we introduce EgoTempo, a dataset specifically designed to evaluate temporal understanding in the egocentric domain. EgoTempo emphasizes tasks that require integrating information across the entire video, ensuring that models would need to rely on temporal patterns rather than static cues or pre-existing knowledge. Extensive experiments on EgoTempo show that current MLLMs still fall short in temporal reasoning on egocentric videos, and thus we hope EgoTempo will catalyze new research in the field and inspire models that better capture the complexity of temporal dynamics. Dataset and code are available at https://github.com/google-research-datasets/egotempo.git.
View details
XR Blocks: Accelerating Human-Centered AI + XR Innovation
Nels Numan
Evgenii Alekseev
Alex Cooper
Min Xia
Scott Chung
Jeremy Nelson
Xiuxiu Yuan
Jolica Dias
Tim Bettridge
Benjamin Hersh
Michelle Huynh
Konrad Piascik
Ricardo Cabello
Google, XR, XR Labs (2025)
Preview abstract
We are on the cusp where Artificial Intelligence (AI) and Extended Reality (XR) are converging to unlock new paradigms of interactive computing. However, a significant gap exists between the ecosystems of these two fields: while AI research and development is accelerated by mature frameworks like PyTorch and benchmarks like LMArena, prototyping novel AI-driven XR interactions remains a high-friction process, often requiring practitioners to manually integrate disparate, low-level systems for perception, rendering, and interaction. To bridge this gap, we present XR Blocks, a cross-platform framework designed to accelerate human-centered AI + XR innovation. XR Blocks provides a modular architecture with plug-and-play components for core abstraction in AI + XR: user, world, peers; interface, context, and agents. Crucially, it is designed with the mission of "minimum code from idea to reality", accelerating rapid prototyping of complex AI + XR apps. Built upon accessible technologies (WebXR, three.js, TensorFlow, Gemini), our toolkit lowers the barrier to entry for XR creators. We demonstrate its utility through a set of open-source templates, samples, and advanced demos, empowering the community to quickly move from concept to interactive prototype.
View details
Preview abstract
Integrating product catalogs and user behavior into LLMs can enhance recommendations with broad world knowledge, but the scale of real-world item catalogs, often containing millions of discrete item identifiers (Item IDs), poses a significant challenge. This contrasts with the smaller, tokenized text vocabularies typically used in LLMs. The predominant view within the LLM-based recommendation literature is that it is infeasible to treat item ids as a first class citizen in the LLM and instead some sort of tokenization of an item into multiple tokens is required. However, this creates a key practical bottleneck in serving these models for real-time low-latency applications.
Our paper challenges this predominant practice and integrates item ids as first class citizens into the LLM. We provide simple, yet highly effective, novel training and inference modifications that enable single-token representations of items and single-step decoding. Our method shows improvements in recommendation quality (Recall and NDCG) over existing techniques on the Amazon shopping datasets while significantly improving inference efficiency by 5x-14x. Our work offers an efficiency perspective distinct from that of other popular approaches within LLM-based recommendation, potentially inspiring further research and opening up a new direction for integrating IDs into LLMs. Our code is available here https://drive.google.com/file/d/1cUMj37rV0Z1bCWMdhQ6i4q4eTRQLURtC/edit
View details
ESAM++: Efficient Online 3D Perception on the Edge
Qin Liu
Lavisha Aggarwal
Vikas Bahirwani
Lin Li
Aleksander Holynski
Saptarashmi Bandyopadhyay
Zhengyang Shen
Marc Niethammer
Ehsan Adeli
Andrea Colaco
2025
Preview abstract
Online 3D scene perception in real time is critical for robotics, AR/VR, and autonomous systems, particularly in edge computing scenarios where computational resources are limited. Recent state-of-the-art methods like EmbodiedSAM (ESAM) demonstrate the promise of online 3D perception by leveraging the 2D visual foundation model (VFM) with efficient 3D query lifting and merging. However, ESAM depends on a computationally expensive sparse 3D U-Net for point cloud feature extraction, which we identify as the primary efficiency bottleneck. In this paper, we propose a lightweight and scalable alternative for online 3D scene perception tailored to edge devices. Our method introduces a 3D Sparse FeaturePyramid Network (SFPN) that efficiently captures multi-scale geometric features from streaming 3D point clouds while significantly reducing computational over-head and model size. We evaluate our approach on four challenging segmentation benchmarks—ScanNet, ScanNet200, SceneNN, and 3RScan—demonstrating that our model achieves competitive accuracy with up to 3×faster inference and 3×small model size compared to ESAM, enabling practical deployment in real-world edge scenarios. Code and models will be released.
View details
Crosslingual Capabilities and Knowledge Barriers in Multilingual Large Language Models
Lynn Chua
Yangsibo Huang
Pritish Kamath
Amer Sinha
Chulin Xie
Chiyuan Zhang
COLM (2025)
Preview abstract
Large language models (LLMs) are typically multilingual due to pretraining on diverse multilingual corpora. But can these models relate corresponding concepts across languages, i.e., be crosslingual? This study evaluates state-of-the-art LLMs on inherently crosslingual tasks. We observe that while these models show promising surface-level crosslingual abilities on machine translation and embedding space analyses, they struggle with deeper crosslingual knowledge transfer, revealing a crosslingual knowledge barrier in both general (MMLU benchmark) and domain-specific (Harry Potter quiz and TOFU benchmark) contexts. Since simple inference-time mitigation methods offer only limited improvement, we propose fine-tuning of LLMs on mixed-language data, which effectively reduces these gaps, even when using out-of-domain datasets like WikiText. Our findings suggest the need for explicit optimization to unlock the full crosslingual potential of LLMs. Our code is available at https://github.com/google-research/crosslingual-knowledge-barriers.
View details
The Cost of Consistency: Submodular Maximization with Constant Recourse
Paul Duetting
Federico Fusco
Ashkan Norouzi Fard
Ola Svensson
Proceedings of the 57th Annual ACM Symposium on Theory of Computing (2025), 1406–1417
Preview abstract
In this work, we study online submodular maximization and how the requirement of maintaining a stable solution impacts the approximation. In particular, we seek bounds on the best-possible
approximation ratio that is attainable when the algorithm is allowed to make, at most, a constant number of updates per step. We show a tight information-theoretic bound of $2/3$ for general monotone submodular functions and an improved (also tight) bound of $3/4$ for coverage functions. Since both these bounds are attained by non poly-time algorithms, we also give a poly-time randomized algorithm that achieves a $0.51$-approximation. Combined with an
information-theoretic hardness of $1/2$ for deterministic algorithms from prior work, our work thus shows a separation between deterministic and randomized algorithms, both information theoretically and for poly-time algorithms.
View details
The Anatomy of a Personal Health Agent
Ahmed Metwally
Ken Gu
Jiening Zhan
Kumar Ayush
Hong Yu
Amy Lee
Qian He
Zhihan Zhang
Isaac Galatzer-Levy
Xavi Prieto
Andrew Barakat
Ben Graef
Yuzhe Yang
Daniel McDuff
Brent Winslow
Shwetak Patel
Girish Narayanswamy
Conor Heneghan
Max Xu
Jacqueline Shreibati
Mark Malhotra
Orson Xu
Tim Althoff
Tony Faranesh
Nova Hammerquist
Vidya Srinivas
arXiv (2025)
Preview abstract
Health is a fundamental pillar of human wellness, and the rapid advancements in large language models (LLMs) have driven the development of a new generation of health agents. However, the solution to fulfill diverse needs from individuals in daily non-clinical settings is underexplored. In this work, we aim to build a comprehensive personal health assistant that is able to reason about multimodal data from everyday consumer devices and personal health records. To understand end users’ needs when interacting with such an assistant, we conducted an in-depth analysis of query data from users, alongside qualitative insights from users and experts gathered through a user-centered design process. Based on these findings, we identified three major categories of consumer health needs, each of which is supported by a specialist subagent: (1) a data science agent that analyzes both personal and population-level time-series wearable and health record data to provide numerical health insights, (2) a health domain expert agent that integrates users’ health and contextual data to generate accurate, personalized insights based on medical and contextual user knowledge, and (3) a health coach agent that synthesizes data insights, drives multi-turn user interactions and interactive goal setting, guiding users using a specified psychological strategy and tracking users’ progress. Furthermore, we propose and develop a multi-agent framework, Personal Health Insight Agent Team (PHIAT), that enables dynamic, personalized interactions to address individual health needs. To evaluate these individual agents and the multi-agent system, we develop a set of N benchmark tasks and conduct both automated and human evaluations, involving 100’s of hours of evaluation from health experts, and 100’s of hours of evaluation from end-users. Our work establishes a strong foundation towards the vision of a personal health assistant accessible to everyone in the future and represents the most comprehensive evaluation of a consumer AI health agent to date.
View details
Responsible Innovation, Moral Imagination, and Epistemic Trust: A Practitioner Perspective
Susan Rubin
Carmen Bush
Geoff Keeling
Ben Zevenbergen
Benjamin Lange
Currently under review - Journal of Responsible Innovation (2025) (to appear)
Preview abstract
This paper is a practitioner's account of Google's Moral Imagination Program for responsible innovation. Elements of the methodology have previously been externalised, e.g
Lange, B., Keeling, G., McCroskery, A., Zevenbergen, B., Blascovich, S., Pedersen, K., ... & Agüera y Arcas, B. (2023). Engaging engineering teams through moral imagination: a bottom-up approach for responsible innovation and ethical culture change in technology companies. AI and Ethics, 1-10. (https://link.springer.com/article/10.1007/s43681-023-00381-7)
Keeling, G., Lange, B., McCroskery, A., Pedersen, K., Weinberger, D., & Zevenbergen, B. (2024). Moral Imagination for Engineering Teams: The Technomoral Scenario. The International Review of Information Ethics, 34(1).(https://informationethics.ca/index.php/irie/article/view/527)
This paper builds on that prior work, intending to help practitioners at other tech companies create their own initiatives that draw broadly on the same ideas.
Abstract: Integrating ethical considerations into technology development can be difficult. It requires an approach that challenges pervasive engineering paradigms and incorporates social non-technical values. We argue that effective ethics interventions in technology development require epistemic trust in the intervening discipline, the facilitator(s), and the content. We base this on our experience in developing, implementing, and iterating the Moral Imagination programme at Google through 70 workshops over four years with a wide variety of teams. We detail best-practices of the moral imagination methodology and how they are conducive toward promoting epistemic trust support. The aim of our discussion is to inform ongoing responsible innovation research alongside professional practice by providing a practitioner’s perspective from the ground.
View details
Preview abstract
Online scams are a growing threat in India, impacting millions and causing substantial financial losses year over year. This white paper presents ShieldUp!, a novel mobile game prototype designed to inoculate users against common online scams by leveraging the principles of psychological inoculation theory. ShieldUp! exposes users to weakened versions of manipulation tactics frequently used by scammers, and teaches them to recognize and pre-emptively refute these techniques. A randomized controlled trial (RCT) with 3,000 participants in India was conducted to evaluate the game's efficacy in helping users better identify scams scenarios. Participants were assigned to one of three groups: the ShieldUp! group (play time: 15 min), a general scam awareness group (watching videos and reading tips for 10-15 min), and a control group (plays "Chrome Dino", an unrelated game, for 10 minutes). Scam discernment ability was measured using a newly developed Scam Discernment Ability Test (SDAT-10) before the intervention, immediately after, and at a 21-day follow-up. Results indicated that participants who played ShieldUp! showed a significant improvement in their ability to identify scams compared to both control groups, and this improvement was maintained at follow-up. Importantly, while both interventions initially led users to to show increased skepticism towards even genuine online offers (NOT Scam scenarios), this effect dissipated after 21 days, suggesting no long-term negative impact on user trust. This study demonstrates the potential of game-based inoculation as a scalable and effective scam prevention strategy, offering valuable insights for product design, policy interventions, and future research, including the need for longitudinal studies and cross-cultural adaptations.
View details
EQuARX: Efficient Quantized AllReduce in XLA for Distributed Machine Learning Acceleration
Preview
Ibrahim Ahmed
Clemens Schaefer
Denis Vnukov
Zenong Zhang
Toli Yevtushenko
2025
Preview abstract
The Latin script is often used informally to write languages with non-Latin native scripts. In many cases (e.g., most languages in India), there is no orthography, meaning that there is no conventional spelling of words in the Latin script, hence there will be high spelling variability in written text. Such romanization can render languages that are normally easily distinguished based on script highly confusable, such as Hindi and Urdu. In this work, we present methods to improve language identification of romanized text by improving methods to synthesize training sets. We find that training on synthetic samples which incorporate natural spelling variation yields higher language identification system accuracy than including available naturally occurring examples in the training set or even training higher capacity models. We demonstrate new state-of-the-art language identification performance on romanized text from 20 Indic languages in the Bhasha-Abhijnaanam evaluation set (Madhani et al., 2023a), improving test F1 from the reported 74.7% (using a pretrained neural model) to 85.4% using a linear classifier trained solely on synthetic data and 88.2% when also training on available harvested text.
View details
Preview abstract
Differentially private (DP) synthetic data is a versatile tool for enabling the analysis of private data. With the rise of foundation models, a number of new synthetic data algorithms privately finetune the weights of foundation models to improve over existing approaches to generating private synthetic data. In this work, we propose two algorithms for using API access only to generate DP tabular synthetic data. We extend the Private Evolution algorithm \citep{lin2023differentially, xie2024differentially} to the tabular data domain, define a workload-based distance measure, and propose a family of algorithms that use one-shot API access to LLMs.
View details
Amplifying Trans and Nonbinary Voices: A Community-Centred Harm Taxonomy for LLMs
Eddie Ungless
Beka Gulotta
Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (2025)
Preview abstract
We explore large language model (LLM) responses that may negatively impact the transgender and nonbinary (TGNB) community and introduce the Transing Transformers Toolkit, T3, which provides resources for identifying such harmful response behaviors. The heart of T3 is a community-centred taxonomy of harms, developed in collaboration with the TGNB community, which we complement with, amongst other guidance, suggested heuristics for evaluation. To develop the taxonomy, we adopted a multi-method approach that included surveys and focus groups with community experts. The contribution highlights the importance of community-centred approaches in mitigating harm, and outlines pathways for LLM developers to improve how their models handle TGNB-related topics.
View details
