Customer-obsessed science
-
June 07, 2024Although work involving large language models predominates, classical and more-general techniques remain well represented.
-
May 31, 2024Novel loss term that can be added to any loss function regularizes interclass and intraclass distances.
-
May 17, 2024A novel loss function and a way to aggregate multimodal input data are key to dramatic improvements on some test data.
-
-
June 9 - 14, 2024
-
June 16 - 21, 2024
-
June 17 - 21, 2024
-
June 03, 2024
A combination of generative AI and computer vision imaging tunnels is helping Amazon proactively improve the customer experience.
-
2024Mitigating hallucinations in large vision-language mod-els (LVLMs) remains an open problem. Recent benchmarks do not address hallucinations in open-ended free-form re-sponses, which we term “Type I hallucinations”. Instead, they focus on hallucinations responding to very specific question formats—typically a multiple-choice response re-garding a particular object or attribute—which we term “Type II hallucinations
-
International Journal of Computer Vision2024Matching algorithms predict relationships between items in a collection. For example, in 1:1 face verification, a matching algorithm predicts whether two face images depict the same per-son. Accurately assessing the uncertainty of the error rates of such algorithms can be challenging when test data are dependent and error rates are low, two aspects that have been often over-looked in the literature. In
-
EUSIPCO 20242024Loudspeaker array beamforming is a common sig-nal processing technique for acoustic directivity control and robust audio reproduction. Unlike their microphone counterpart, loudspeaker constraints are often heterogeneous due to arrayed transducers with varying operating ranges in frequency, acoustic-electrical sensitivity, efficiency, and directivity. This work pro-poses a frequency-regularization method
-
2024Multi-task learning (MTL) considers learning a joint model for multiple tasks by optimizing a convex combination of all task losses. To solve the optimization problem, existing methods use an adaptive weight updating scheme, where task weights are dynamically adjusted based on their respective losses to prioritize difficult tasks. How-ever, these algorithms face a great challenge when-ever label noise is
-
In the dynamic marketplace, vendors continuously seek innovative ideas for new products and ways to improve existing ones. These ideas can be uncovered by analyzing text data, such as product descriptions and customer reviews. However, the ever-increasing volume of text data poses a challenge in extracting meaningful insights. Therefore, this study addresses the challenge of extracting actionable insights
News and features
-
April 26, 2024Awardees, who represent 51 universities in 15 countries, have access to Amazon public datasets, along with AWS AI/ML services and tools.
-
April 09, 2024How the team behind Echo Frames delivered longer battery life and improved sound quality inside the slim form factor of a pair of eyeglasses.
-
March 21, 2024The principal economist and his team address unique challenges using techniques at the intersection of microeconomics, statistics, and machine learning.