To transform scientific domains, foundation models will require physical-constraint satisfaction, uncertainty quantification, and specialized forecasting techniques that overcome data scarcity while maintaining scientific rigor.
The language AI agents might speak, sharing context without compromising privacy, modeling agentic negotiations, and understanding users’ commonsense policies are some of the open scientific questions that researchers in agentic AI will need to grapple with.
With a novel parallel-computing architecture, a CAD-to-USD pipeline, and the use of OpenUSD as ground truth, a new simulator can explore hundreds of sensor configurations in the time it takes to test just a few physical setups.
From reimagining storage to serverless computing, Aurora continues to push the boundaries of what's possible in database technology.
Amazon VP and distinguished scientist Byron Cook explains how AWS's new Automated Reasoning checks address key challenges in automated reasoning: translating natural to structured language, defining truth, and definitive reasoning.
Customer-obsessed science
Research areas
-
September 2, 2025Audible's ML algorithms connect users directly to relevant titles, reducing the number of purchase steps for millions of daily users. -
Featured news
University teams battle to harden and hack AI coding assistants in head-to-head tournament
David Chang/Getty Images/iStockphoto
Led by David Luan and Pieter Abbeel, the lab will focus on developing new foundational capabilities for enabling useful AI agents.
The company's new state-of-the-art foundation models deliver frontier intelligence and industry-leading price performance.
-
Suggesting relevant questions to users is an important task in various applications, such as community Q&A or e-commerce websites. To ensure that there is no redundancy in the selected set of candidate questions, it is essential to filter out any near-duplicate questions. Identifying near-duplicate questions has another use case in light of the adoption of Large Language Models (LLMs) – fetching pre-computed
-
IEEE Robotics and Automation Letters 2024, IROS 20242024In this paper we propose an approach to trajectory planning based on the purpose of the task. For a redundant manipulator, many end effector poses in the task space can be achieved with multiple joint configurations. In planning the motion, we are free to choose the configuration that is optimal for the particular task requirement. Many previous motion-planning approaches have been proposed for the sole
-
In the field of Natural Language Processing (NLP), sentence pair classification is important in various real-world applications. Bi-encoders are commonly used to address these problems due to their low-latency requirements, and their ability to act as effective retrievers. However, bi-encoders often under-perform compared to cross-encoders by a significant margin. To address this gap, many Knowledge Distillation
-
Federated Learning (FL) is a popular algorithm to train ma-chine learning models on user data constrained to edge devices (for example, mobile phones) due to privacy concerns. Typically, FL is trained with the assumption that no part of the user data can be egressed from the edge. However, in many production settings, specific data-modalities/meta-data are limited to be on device while others are not. For
-
AISTATS 20242024Multi-objective optimization (MOO) aims to optimize multiple, possibly conflicting objectives with widespread applications. We introduce a novel interacting particle method for MOO inspired by molecular dynamics simulations. Our approach combines overdamped Langevin and birth-death dynamics, incorporating a “dominance potential” to steer particles toward global Pareto optimality. In contrast to previous
Conferences
Academia
View allWhether you're a faculty member or student, there are number of ways you can engage with Amazon.
View all
Gretchen Ertl
The program offers unrestricted funds and other resources to support research at academic institutions and non-profit organizations in areas that align with our mission.
A global university competition to drive secure innovation in generative AI technology, which focuses on responsible AI and large language model coding security.
Credit: Wolfram Scheible
We partner with particular academic organizations across the world for deep and sustained collaborations in multiple research areas of mutual interest.
Courtesy of Pai-Ling Yin
We hire world-class academics to work on large-scale technical challenges, while they continue to teach and conduct research at their universities.
