About
Research & Development
PolySim: A New Framework for Robust Humanoid Robot Control Across Diverse Simulations
TLDR: PolySim is a novel training platform that addresses the "sim-to-real gap" in humanoid robot control by training policies across multiple, heterogeneous simulators simultaneously....
Research & Development
PyramidStyler: Advancing Neural Style Transfer with Multi-Scale Encoding
TLDR: PyramidStyler is a new transformer-based neural style transfer model that addresses efficiency and quality issues in artistic image synthesis. It introduces Pyramidal Positional...
Research & Development
Advancing Lifelong Learning for Large Language Models with REPAIR
TLDR: REPAIR is a novel framework for editing large language models (LLMs) that enables precise, low-cost updates while preserving existing knowledge. It tackles challenges...
Research & Development
A Unified Approach to Early Disease Diagnosis with Multimodal AI
TLDR: The research proposes a multimodal foundation model using an attention-based transformer framework to integrate diverse patient data (electronic health records, medical imaging, genetics,...
Research & Development
Verifying AI Image Origin: A New Approach with ZK-SNARKs
TLDR: ZK-WAGON is a novel system that uses Zero-Knowledge Succinct Non-Interactive Arguments of Knowledge (ZK-SNARKs) to imperceptibly watermark images generated by AI models like...
Research & Development
Enhancing Video Anomaly Detection with Action-Focused Prompts for AI Models
TLDR: ASK-HINT is a new method that uses detailed, action-specific questions to guide AI vision-language models in detecting unusual events in videos. Unlike abstract...