A new hybrid optimization approach allows edge devices to fine-tune vision-language models using only forward passes, achieving up to 7% higher accuracy than existing techniques.
"Reinforcement learning gyms" train agents on the many low-level tasks that they must chain together to execute customer requests.
New audio-processing technology is making entertainment more accessible for millions of viewers.
Reasoning models can generate seven to 10 times as many tokens as necessary on simple tasks, creating unsustainable costs at scale. Amazon's vision for metacognitive AI could fundamentally shift how models allocate computational resources.
Customer-obsessed science
Research areas
-
January 13, 20267 min readLeveraging existing environment simulators and reward functions based on verifiable ground truth boosts task success rate, even with small models and small training datasets. -
December 29, 20256 min read -
December 29, 20259 min read
-
December 8, 20258 min read
-
December 5, 20256 min read
Featured news
Hosted by Dr. Danielle Perszyk, cognitive scientist at Amazon's AGI Lab, the podcast features conversations with leading AI researchers about the breakthroughs needed to achieve general intelligence.
Challenge pushes teams to demonstrate measurable gains in secure-coding performance while building AI agents that advance real-world utility and reliability at scale.
Meet the 63 Amazon Research Award (ARA) recipients, who represent 41 universities in 8 countries.
Initiative will fund over 100 doctoral students researching machine learning, computer vision, and natural-language processing at nine universities.
-
2025Fixed-size learned representations (dense representations, or embeddings) are widely used in many machine learning applications across language, vision or speech modalities. This paper investigates the role of the temperature parameter in contrastive training for text embeddings. We shed light on the impact this parameter has on the intrinsic dimensionality of the embedding spaces obtained, and show that
-
2025Direct alignment algorithms (DAAs), such as direct preference optimization (DPO), have become popular alternatives for Reinforcement Learning from Human Feedback (RLHF) due to their simplicity, efficiency, and stability. However, the preferences used in DAAs are usually collected before the alignment training begins and remain unchanged (off-policy). This design leads to two problems where the policy model
-
2025Generative models that satisfy hard constraints are critical in many scientific and engineering applications, where physical laws or system requirements must be strictly respected. Many existing constrained generative models, especially those developed for computer vision, rely heavily on gradient information, which is often sparse or computationally expensive in some fields, e.g., partial differential
-
2025Vision-Language Models (VLMs) have demonstrated impressive performance across a versatile set of tasks. A key challenge in accelerating VLMs is storing and accessing the large Key-Value (KV) cache that encodes long visual contexts, such as images or videos. While existing KV cache compression methods are effective for Large Language Models (LLMs), directly migrating them to VLMs yields suboptimal accuracy
-
2025Faithfulness evaluators based on large language models (LLMs) are often fooled by the fluency of the text and struggle with identifying errors in the summaries. We propose an approach to summary faithfulness evaluation in which multiple LLM-based agents are assigned initial stances (regardless of what their belief might be) and forced to come up with a reason to justify the imposed belief, thus engaging
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.
