About
Research & Development
Advancing Realistic AI-Generated Co-Speech Gesture Videos with a Public Dataset
TLDR: A new research paper introduces a lightweight framework for generating high-fidelity, audio-aligned co-speech gesture videos using 2D full-body skeletons and a diffusion model....
Research & Development
Unlocking Better Federated Learning with LIVAR’s Dual Aggregation Strategy
TLDR: LIVAR is a new Federated Learning framework that improves model aggregation by using naturally occurring training signals. It combines client-specific classification heads based...
Research & Development
Crafting Better LLM Responses: A New Self-Review Framework for Instruction Following
TLDR: Re5 is a self-evaluation and revision framework for Large Language Models (LLMs) that significantly improves their ability to follow complex instructions while maintaining...
Research & Development
HeLo: Advancing Emotion Recognition with Multi-Modal Data and Label Connections
TLDR: HeLo is a new framework for emotion distribution learning that addresses challenges in multi-modal emotion recognition. It effectively fuses heterogeneous physiological and behavioral...
Research & Development
OpenDPDv2: Enhancing Efficiency in Neural Network Digital Predistortion for Wireless Systems
TLDR: OpenDPDv2 is a new framework for Digital Predistortion (DPD) in wireless systems, addressing the high power consumption of neural network-based DPDs. It introduces...
News & Current Events
Ririe School District Initiates AI Policy Development, Aligns with ISBA Guidance
TLDR: The Ririe School District is actively working on establishing comprehensive guidelines for the use of Artificial Intelligence within its schools. This initiative follows...