New research indicates that AI agents learn compressible models of data, which don’t have enough space to enable memorization.
How the Verus "program verifier", which automatically checks code against a mathematical specification of its functionality, helps increase security assurance in software projects.
Ten years after we founded the Automated Reasoning Group, mathematical logic has moved from academic research into production services that secure millions of customer workloads — demonstrating that systems can be provably correct, not just probably correct.
PatientAgentBench generates a synthetic patient health record, a realistic clinical vignette, and a patient agent that converses with the AI system under evaluation, to capture what a patient-facing agent actually has to do.
A new chiplet architecture, custom die-to-die connectivity, and support for DDR5-8800 memory and the latest PCIe gen6 interconnects improve performance by 25% for general-purpose and agentic AI workloads.
Customer-obsessed science
Research areas
-
September 25, 202612 min readUsing the Neuron Kernel Interface, a Reactor–AWS collaboration tackled the dynamic shapes, memory access patterns, and cache management that make real-time autoregressive diffusion hard—building techniques that generalize across models. -
September 21, 202611 min read -
-
August 21, 20269 min read
-
July 30, 20268 min read
Featured news
The Build on Trainium program is a $110 million credit initiative supporting AI research at 30 universities including Stanford, UC Berkeley, UIUC, UCLA, CMU, and MIT, with a focus on Responsible AI.
The Antibody Developability Benchmark is powered by one of the most diverse antibody datasets in public literature, enabling transparent performance evaluation for AI-guided antibody design.
Challenge pushes teams to demonstrate measurable gains in secure-coding performance while building AI agents that advance real-world utility and reliability at scale.
Initiative will fund over 100 doctoral students researching machine learning, computer vision, and natural-language processing at nine universities.
-
COLM 2026 Workshop on Agent Behavior2026Single-metric task-success scores systematically misrank LLM agents in domain-specific deployments: two configurations can achieve identical end-to-end accuracy while exhibiting structurally different production behaviors — silent argument hallucination, infinite tool loops, redundant retries, semantically-equivalent-but-syntactically-divergent queries. We propose a behavioral evaluation protocol that decomposes
-
ACM SIGSPATIAL 20262026Accurate map paths are crucial for routing vehicles, pedestrians, and autonomous systems, yet keeping that data fresh demands continual editing of features and geometry, and labeling of attributes such as surface type and lane count. At scale this is labor-intensive, and each change calls for spatial reasoning and verification that has resisted full automation, keeping a human in the loop. We present MapScout
-
ACM SIGSPATIAL 20262026Last-mile logistics efficiency largely depends on map data being accurate, fresh, and having high geospatial coverage. Operators must preserve these qualities when migrating between heterogeneous map data providers. Existing work in the literature addresses road-geometry conflation; however, the conflation of road network attributes and their transition relations remains under-studied, especially at a production-level
-
CIKM 20262026Production search indices operate under strict storage and latency budgets that prevent indexing the full corpus (e.g.: web documents). When only a fraction of documents can be retained, the system must decide which ones to keep. Existing pruning methods make this decision using query-agnostic signals, including lexical quality, embedding magnitude, and corpus centrality or single-feature query-aware heuristics
-
RecSys 20262026Off-policy evaluation ( OPE) estimates the performance of new recommendation policies using logged data, thus enabling fast, safe and inexpensive iteration prior to costly A/B tests. To evaluate ranking policies, existing OPE estimators all make structural assumptions about user behavior, leading to a spectrum of trade-offs between bias and variance. The recently proposed INTERPOL estimator navigates these
Collaborations
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.
