Search Results

MemX virtualizes cluster-wide memory into a shared, low-latency resource pool for virtualized environments, enabling applications to access datasets larger than the memory capacity of a single machine. By aggregating distributed memory resources and providing transparent remote memory access, MemX allows cloud providers and data center operators to...

This technology enables Vision Language Models to interpret raw floor plan images and generate accurate navigation plans without requiring structured digital maps. By combining visual and text-based reasoning, it allows robots and smart devices to understand spaces directly from images. The result is flexible, scalable navigation that works wherever...

This technology enables high-quality real-time 360-degree video communication by adapting video content to each user’s viewport. By concentrating resolution where the user is actually looking, it reduces bandwidth waste and improves visual clarity under tight latency constraints. The result is an efficient, immersive communication experience optimized...

This technology allows servers to be updated without downtime, solving the common problem that updating hypervisors or operating systems usually requires reboots or temporarily stopping applications. Instead of copying large amounts of memory, it switches memory control in place, enabling updates in under a second while applications keep running. Background:...

A Bernoulli Process Topic (BPT) model that enables knowledge discovery from citation networks to improve data mining tasks Background: Knowledge discovery from citation networks (i.e., textual data with links such as scientific articles, legal documents, webpages, and emails) provides insight into vast areas since huge repositories are made available...

This technology provides a secure virtual machine checkpointing mechanism that prevents sensitive data from being unintentionally stored in system snapshots. By selectively excluding confidential processes and their memory contents, it reduces the risk of data exposure during backup and recovery. The approach enhances privacy without significantly impacting...

Semiparametric regularization based approach allows a family of algorithms to be developed based on various choices of the original RKHS and the loss function Technology Overview: Labeled data are often expensive to obtain since they require the efforts of experienced experts. Meanwhile, the unlabeled data are relatively easy to collect. Semiparametric...

This technology introduces a faster and more scalable machine learning framework for analyzing and retrieving multimodal data such as images and text. It enables real-time querying and pattern discovery even in very large databases while maintaining high accuracy in complex data environments. The approach significantly improves efficiency for large-scale...

This technology introduces a probabilistic framework that clusters complex, interconnected data by modeling relationships across multiple object types. It enables more accurate pattern discovery by capturing both individual attributes and interactions simultaneously. The approach unifies multiple clustering methods into a single flexible system, supporting...

A novel general algorithm to cluster multi-type interrelated data objects simultaneously by iteratively embedding each type of data objects into low dimensional spaces Background: Most clustering approaches in the literature focus on "flat" data in which each data object is represented as a fixed-length feature vector. However, many real-world...