Customer-obsessed science
Research areas
-
July 10, 20265 min readHydroShear, 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.
-
July 9, 202610 min read
-
-
Featured news
-
IGARSS 20262026Remote sensing imagery is a a rich source that serves a wide range of applications including urban planning, land management, environmental monitoring, and digital map enrichment by detecting roads and building outlines. Nevertheless, occlusions from trees often hide important features like roads from bird's-eye viewpoint. To tackle this problem, we propose GeoRoadInpaint model, leveraging stable diffusion
-
ICML 2026 Workshop on Generative and Agentic AI for Biology2026Therapeutic antibody discovery remains slow and resource-intensive, with traditional methods providing limited control over epitope selection. We present a workflow for de novo nanobody design applied to a novel Desmoplastic Small Round Cell Tumor target encompassing four stages: (1) epitope identification guided by our hotspot recommendation agent using physical chemistry-based structure and sequence analysis
-
ISACE 20262026Agentic AI systems can access vast data but struggle to apply domain expertise, namely the contextual understanding of how to use specialized information. This paper presents a practical framework for encoding such expertise, demonstrated with the National Football League (NFL) through NFL Fantasy AI, a production system delivering analyst-grade fantasy football advice, as assessed by NFL Pro analysts.
-
CVPR 2026 EarthVision Workshop2026Building outline extraction from remote sensing imagery traditionally relies on segmentation or detection followed by post-processing to derive polygonal geometries. Despite advances in sequential prediction methods [2, 20], end-to-end extraction remains challenging, often missing buildings or requiring additional refinement steps. In this work, we reformulate building outline extraction as next-coordinate
-
ICLR 2026 Workshop on AI with Recursive Self-Improvement2026Foundation-model upgrades frequently break deployed prompt-based systems: target models differ in chat-template conventions, multimodal interfaces, context limits, and structured-output reliability. We study cross-model prompt adaptation: given a prompt program validated on a source model, produce a target-model prompt that preserves a semantic contract and an interface contract under bounded regression
Collaborations
View allWhether you're a faculty member or student, there are number of ways you can engage with Amazon.
View all