Opening new paths in aging research
Calico Life Sciences uses Co-Scientist to connect scattered findings and generate new leads in aging research.
#Research
#Research
0 todayMay 16
Uncovering repurposed medicines to fight liver fibrosis
Stanford geneticist uses Co-Scientist to help find new treatments for chronic liver disease and liver fibrosis.
May 12
Co-Scientist: A multi-agent AI partner to accelerate research
Introducing Co-Scientist, a collaborative AI partner built with Gemini to help researchers accelerate scientific breakthroughs.
Apr 30
Enabling a new model for healthcare with AI co-clinician
Researching the path to AI-augmented care and development of an AI co-clinician.
Four ways Google Research scientists have been using Empirical Research Assistance
Google Research described how scientists use Empirical Research Assistance (ERA) to advance research in four areas: epidemic forecasting, cosmology, carbon monitoring and neuroscience.
Apr 22
Decoupled DiLoCo: A new frontier for resilient, distributed AI training
Google DeepMind published a paper proposing Decoupled DiLoCo, splitting large-scale training into decoupled compute islands that communicate through asynchronous data streams, isolating local hardware failures.
ReasoningBank: Enabling agents to learn from experience
Google Research released the ICLR paper ReasoningBank, an open-source agent memory framework that distils high-level reasoning patterns from successful and failed experiences.
Apr 20
Gradient-based Planning for World Models at Longer Horizons
GRASP is a new gradient-based planner for learned dynamics (a “world model”) that makes long-horizon planning practical by (1) lifting the trajectory into virtual states so optimization is parallel across time, (2) adding stochasticity directly to the state iterates for exploration, and (3) reshaping gradients so actions get clean signals while we avoid brittle “state-input” gradients through high-dimensional vision models.
Apr 16
Designing synthetic datasets for the real world: Mechanism design and reasoning from first principles
Google proposed Simula, a reasoning-first synthetic data framework reframing dataset construction as mechanism design. It builds datasets from scratch without seed data through global diversification, local diversification, complexification and dual-critic quality checks.
AI-generated synthetic neurons speed up brain mapping
Google Research proposed MoGen (Neuronal Morphology Generation), using PointInfinity point-cloud flow matching to generate synthetic neuronal shapes and supplement PATHFINDER reconstruction training. It reduced reconstruction error by 4.4% on held-out mouse axons, mainly by reducing merge errors.
Apr 14
Towards developing future-ready skills with generative AI
Google released Vantage, a research experiment using generative AI simulated conversations to assess future skills such as problem-solving and collaboration in school and university students, with English registration open on Google Labs. An Executive LLM dynamically guides dialogue and an AI Evaluator scores against rubrics.
Apr 9
ConvApparel: Measuring and bridging the realism gap in user simulators
Google Research released ConvApparel, with over 4,000 human–machine multi-turn clothing-shopping conversations, nearly 15,000 turns, to quantify realism gaps between LLM user simulators and real people.
Improving the academic workflow: Introducing two AI agents for better figures and peer review
Google released PaperVizAgent for paper illustrations and ScholarPeer for automated peer review. PaperVizAgent combines retrieval, planning, style, visualisation and review agents, scoring 60.2 on a 0–100 evaluation, ahead of GPT-Image-1.5, Nano-Banana-Pro and Paper2Any. It was the only framework above the 50.0 human baseline.
Apr 3
Evaluating alignment of behavioral dispositions in LLMs
Google Research proposed situational judgement tests (SJT) adapted from psychological questionnaires such as IRI and ERQ to evaluate alignment between LLM behaviours and human consensus in real user–assistant scenarios.
Apr 1
Building better AI benchmarks: How many raters are enough?
Google Research studied trade-offs between evaluation item counts and annotators per item in Forest vs Tree: The (N, K) Trade-off in Reproducible ML Evaluation. Common counts of one, three or five annotators are often insufficient; reflecting human disagreement frequently requires more than ten.
Mar 31
Safeguarding cryptocurrency by disclosing quantum vulnerabilities responsibly
Google Quantum AI published a white paper saying future quantum computers would require fewer qubits and gates than previously estimated to break the elliptic-curve cryptography protecting cryptocurrencies.
Mar 25
TurboQuant: Redefining AI efficiency with extreme compression
Google Research released TurboQuant, compressing the KV cache to three bits without loss of model accuracy or training and fine-tuning, alongside the QJL and PolarQuant methods.
Mapping the modern world: How S2Vec learns the language of our cities
S2Vec uses S2 Geometry partitions and rasterised features to turn buildings and roads into multilayer images, learning general embeddings with masked autoencoding (MAE). It beats SATCLIP and GEOCLIP on zero-shot geographic socioeconomic extrapolation such as US population density and median income, but needs improvement on tree cover and elevation.
Mar 18
Improving breast cancer screening workflows with machine learning
The AIMS collaboration between Google Research and several NHS organisations published two companion studies in Nature Cancer evaluating an AI breast cancer detection system within NHS screening workflows.
Mar 17
Testing LLMs on superconductivity research questions
Google and Cornell published a PNAS study asking GPT-4o, Perplexity, Claude 3.5, Gemini Advanced Pro 1.5, NotebookLM and a custom RAG system 67 expert superconductivity questions. Twelve international experts blindly assessed balance, comprehensiveness, concision, evidence, image relevance and qualitative feedback.
Mar 13
Identifying Interactions at Scale for LLMs
--> Understanding the behavior of complex machine learning systems, particularly Large Language Models (LLMs), is a critical challenge in modern artificial intelligence. Interpretability research aims to make the decision-making process more transparent to model builders and impacted humans, a step toward safer and more trustworthy AI.
Mar 12
Introducing Groundsource: Turning news reports into data with Gemini
Google Research launched Groundsource, using Gemini to extract structured historical disaster data from global news. Its first open dataset contains 2.6 million urban flash-flood records across over 150 countries, spanning 2000 to the present.
Exploring the feasibility of conversational diagnostic AI in a real-world clinical study
Google Research and Beth Israel Deaconess Medical Center conducted a prospective, single-centre feasibility study in which AMIE held text consultations with patients before outpatient primary care visits, under live physician supervision by video.
Jan 10
Information-Driven Design of Imaging Systems
An encoder (optical system) maps objects to noiseless images, which noise corrupts into measurements. Our information estimator uses only these noisy measurements and a noise model to quantify how well measurements distinguish objects. Many imaging systems produce measurements that humans never see or cannot interpret directly. Your smartphone processes raw sensor data through algorithms before producing the final photo.
Nov 1
RL without TD learning
In this post, I’ll introduce a reinforcement learning (RL) algorithm based on an “alternative” paradigm: divide and conquer . Unlike traditional methods, this algorithm is not based on temporal difference (TD) learning (which has scalability challenges ), and scales well to long-horizon tasks. We can do Reinforcement Learning (RL) based on divide and conquer, instead of temporal difference (TD) learning.
Sep 1
What exactly does word2vec learn?
What exactly does word2vec learn, and how? Answering this question amounts to understanding representation learning in a minimal yet interesting language modeling task. Despite the fact that word2vec is a well-known precursor to modern language models, for many years, researchers lacked a quantitative and predictive theory describing its learning process.