Using ensembles of agents to generate and refine interactions annotated with chains of thought improves performance on a battery of benchmarks by an average of 29%.
Language models inspired by specialized processing regions in the brain offer significant time and cost savings.
An ensemble of models, weighted according to their reluctance to flag anomalies, outperforms its predecessors.
Unique end-of-arm tools with three-dimensional force sensors and innovative control algorithms enable robotic arms to “pick” items from and “stow” items in fabric storage pods.
We present Amazon Nova Premier, our most capable multimodal foundation model and teacher for model distillation.
Customer-obsessed science
Research areas
-
August 4, 2025Translating from natural to structured language, defining truth, and definitive reasoning remain topics of central concern in automated reasoning, but Amazon Web Services’ new Automated Reasoning checks help address all of them.
Featured news
Created for students keen to build their career in machine learning, the fifth edition of the program is now open for all eligible students from recognized institutes in India.
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.
-
Query-to-Product Type (Q2PT) is a crucial e-commerce query understanding signal, which directly influences search results relevance and customer UX experience. This imposes high standards on the industrial Q2PT classification models, which have to be regularly monitored for quality among all predicted product types and use cases at scale. Existing solutions for such Q2PT model evaluation involve human-labeled
-
2025Task-oriented Dialog systems (ToD) are essential in automating user interactions, but their complex design and dynamic nature make evaluation particularly challenging. Current evaluation methodologies heavily depend on human annotators, which can be inefficient, subjective, and expensive to scale. To advance the field, there is a pressing need for a reliable, scalable, and systematic evaluation framework
-
2025Constrained decoding with lookahead heuristics (CDLH) is a highly effective method for aligning LLM generations to human preferences. However, the extensive lookahead rollout operations for each generated token makes CDLH prohibitively expensive, resulting in low adoption in practice. In contrast, common decoding strategies such as greedy decoding are extremely efficient, but achieve very low constraint
-
2025Ensuring that large language models (LLMs) do not generate harmful text is critical for their safe deployment. A common failure mode involves producing toxic responses to otherwise innocuous prompts. While various detoxification methods have been proposed, the underlying mechanisms that drive toxic generation in LLMs are not yet fully understood. Our work aims to provide a mechanistic understanding of toxic
-
ICSE 20252025Software developers increasingly rely on AI code generation utilities. To ensure that “good” code is accepted into the code base and “bad” code is rejected, developers must know when to trust an AI suggestion. Understanding how developers build this intuition is crucial to enhancing developer-AI collabo-rative programming. In this paper, we seek to understand how developers (1) define and (2) evaluate the
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.
