About
Research & Development
Achieving Stable 3D Vision in Surgical Videos with Time-Switchable AI Learning
TLDR: TiS-TSL is a new AI framework that solves the problem of unstable 3D depth perception in surgical videos, which traditionally suffer from limited...
Research & Development
Unveiling the Dual Nature of LLM Safety: A New Framework to Bypass Alignment
TLDR: A new research paper introduces Differentiated Bi-Directional Intervention (DBDI), a white-box framework that challenges the traditional view of LLM safety alignment as a...
Research & Development
Task-Aware Explanations Enhance AI Prediction and Understanding
TLDR: ALIGN is a new framework that jointly trains an AI classifier and a masker. Unlike previous methods that rely on noisy manual annotations,...
Research & Development
GEWDiff: A Novel Approach to Hyperspectral Image Super-resolution
TLDR: GEWDiff is a new diffusion model for 4x hyperspectral image super-resolution that addresses challenges like high dimensionality and geometric distortion. It uses a...
Research & Development
Enhancing Controllability and Latent Space Regularization in AI Music Generation with Power Transforms
TLDR: This research paper introduces a novel Power Transform (PT)-based regularization method for deep variational Bayesian models used in attribute-controlled symbolic music generation. It...
Research & Development
Unlocking Adaptive Recommendations: How AdaRec Uses LLMs for Personalized Experiences
TLDR: AdaRec is a new LLM-based framework for adaptive personalized recommendations. It uses "narrative profiling" to convert user-item interactions into natural language, making them...