Amazon Science

amazon.science Imported from official source

Amazon Science — publications from its own official source.

Publications 38

  1. Mathematical proof, powered by AI. And no hallucinations!

    With new tools that make it easier to check mathematical proofs — even as they’re being constructed — AI is becoming a valuable partner in mathematical discovery.

    AI
  2. Using mathematics to validate network configurations

    Static analysis of deployed networks helps reduce the risk that sensitive data leaks across partitions, supporting the security and reliability of global-scale networking.

  3. Mitra-v2: Scaling synthetic pretraining for tabular foundation models

    Expanding the mix of synthetic data used to train a tabular foundation model enables it to adapt to new tables without new models and lead the single-model field on the TabArena benchmark.

    AI
  4. Hamed Karimi wins ECML PKDD Test-of-Time Award

    Amazon senior applied scientist and coauthors showed that an overlooked mathematical condition from 1963 proves why frequently used machine-learning optimization methods work well.

    AI
  5. Speeding up language models for travel applications

    Two optimization techniques, when combined, can significantly improve a model’s speed and cost effectiveness without sacrificing quality.

    AI
  6. Better speech synthesis for audiobooks

    “Controllable text to speech” makes speech more expressive, character voices more distinct and consistent.

    AI
  7. The science behind TorchNeuron, Amazon’s new PyTorch backend for AWS Trainium chips

    Runtime optimizations and a thorough evaluation process driven by agentic AI help ensure that models written in PyTorch will execute efficiently and reliably on Trainium systems.

    AI
  8. Strengthening AI safety evaluation through independent third-party review

    How Amazon partnered with AVERI to independently assess — and improve — our internal safety testing of Amazon models.

    AI
  9. Can we change an AI model’s architecture without retraining it from scratch?

    “Priming” reuses pretrained Transformer weights to build models with more efficient architectures at less than 1% of the original training cost and with throughputs up to twice as high.

    AI
  10. Graph-centric agentic intelligence

    Augmenting a network graph with agentic AI produces a “digital twin” that can help isolate network failures.

    AI
  11. A kernel-centric path to real-time video generation on Trainium

    Using 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 te...

    AI
  12. Amazon launches research initiative with Stanford University to advance AI and science

    The collaboration aims to advance research while broadening participation and translating discovery into real-world solutions.

    AI
  13. Advancing AI for biology: Teaching models to design and characterize antibodies

    Three new papers from Amazon Bio Discovery address bottlenecks in AI-driven antibody engineering, from benchmarking binding predictors to experimentally validating de novo design.

    AI
  14. Why don’t machine learning research agents overfit?

    New research indicates that AI agents learn compressible models of data, which don’t have enough space to enable memorization.

    AI
  15. Developing provably correct Rust code with Verus

    How the Verus "program verifier", which automatically checks code against a mathematical specification of its functionality, helps increase security assurance in software projects.

  16. When LLM judges agree, should we believe them?

    Discounting the opinions of LLM judges with highly correlated outputs ensures that panels of judges reflect a true diversity of perspectives.

    AI
  17. SOP-Bench: A new benchmark for evaluating AI agents on real business procedures

    Extendable framework enables testing agents on the full set of capabilities required to successfully complete a procedure, not isolated proxy tasks.

    AI
  18. A decade of mathematical certainty: Reflections on the Automated Reasoning Group

    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 tha...

    AI
  19. AWS Trainium Frontier competition: Co-design models and kernels on purpose-built AI chips

    A competition with a finalist ceremony during NeurIPS 2026, challenging researchers to train language models from scratch on Trainium, exploring what optimal architectures look like when the hardwa...

    AI
  20. 34 Amazon Research Awards Build on Trainium recipients announced

    Amazon announces 34 recipients of the Build on Trainium program, a $110 million credit initiative supporting AI research at 30 universities including Stanford, UC Berkeley, UIUC, UCLA, CMU, and MIT...

    AI
  21. How controllers from industrial machinery can coordinate multitask machine learning

    Instead of compromising among parameter updates dictated by different training objectives, ControlG allocates computational capacity to objectives sequentially and dynamically.

    AI
  22. A new benchmark for evaluating patient-facing health AI agents

    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-facin...

    AI
  23. Amazon is investing in the Lean Focused Research Organization

    As AI agents take on higher-stakes decisions, Lean programming language makes it possible to mathematically prove they will behave safely.

    AI
  24. Amazon and University of Michigan give robots a sense of touch

    HydroShear, a new physics-based simulator, teaches robots how to use their sense of touch to perform complex manipulation tasks, in a way that transfers seamlessly to the real world.

    AI
  25. Capturing token IDs during agentic interactions for better reinforcement learning

    A new Rust proxy called Turnstile sits between the model backend and the agent harness to capture information lost in mere text transcripts.

    AI
  26. How Amazon tracks carbon intensity across its operations

    Amazon is developing precise, sector-specific approaches to measuring decarbonization progress — starting with emissions per unit shipped.

  27. The fuel of the future is already here: Why TRISO matters

    Millimeter-scale particles of nuclear-reactor fuel are encased in four layers of different materials that act as a “miniature containment system”.

  28. Graviton5’s improved design increases speed and energy efficiency — beyond Moore’s law

    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 wor...

    AI
  29. EC2’s formally verified “isolation engine” provides mathematical assurance of virtual-machine isolation

    Splitting the “separation kernel” off from the rest of the Nitro security system and using only a subset of the Rust programming language to code it enabled its formal verification.

  30. Real-world grounding in agentic AI

    Four approaches can dramatically improve the performance and trustworthiness of AI agents in operational environments.

    AI
  31. Bridging intent and execution in agentic systems

    The harnesses that mediate between models and tools in agentic systems are becoming their own performance bottleneck, but a few simple design principles can fix what ails them.

    AI
  32. Ground truth is a process, not a dataset

    Automatically fact-checking long, AI-generated research reports poses new challenges — including benchmarking.

    AI
  33. How flat is replacing fat in AWS data center networks

    “Quasi-random” network topologies and new passive optical components called ShuffleBoxes make more-efficient flat networks as practical as traditional “fat-tree” networks.

  34. Amazon Research Awards recipients announced

    Awardees represent more than 49 universities in 11 countries. Recipients have access to Amazon public datasets, along with AWS AI/ML services and tools.

  35. Diverse reasoning traces teach LLMs to make better decisions

    How to train language models to generate diverse, accurate reasoning paths using tokens that control distinct reasoning strategies.

    AI
  36. Making LLMs faster without sacrificing accuracy

    A new scaling law that relates particular architectural choices to loss helps identify models that improve throughput by up to 47% with no loss of accuracy.

    AI
  37. Promptimus: Improving already good LLM prompts with zero manual engineering

    By focusing on specific failure points and suggesting targeted solutions, a new automated prompt-engineering framework improves prompt performance without compromising existing functionality.

    AI
  38. Navigating uncertainty in Amazon's middle-mile network

    Amazon engineers and scientists have created new tools to optimize delivery networks under uncertainty — and keep them adapting without missing a beat.

    AI

Officially records where a publication came from, not whether it is true. Imported records are reproduced from an organization's own official source.