Researchers from the USC Viterbi School of Engineering are presenting 24 papers at the 2025 International Conference on Learning Representations (ICLR), Apr. 24-28, one of the premier global ...

The study’s headline finding is clear: artificial intelligence significantly depresses the demand for low-skilled labor, ...

Right where AI struggles: The unpredictable. The emotional. The creative. The nonlinear. What does nonlinear thinking—or “squiggly” thinking, as I like to call it—look like?

To address these issues, we propose a novel HTD method based on diffusion model and convolutional gated linear unit (HTD-DMCG). First, the diffusion model is integrated with MixUp for data ...

Keywords: video generation, text-to-image, diffusion transformer, image generation Accelerating Diffusion Transformers with Dual Feature Caching (Published: 2024-12-25) Authors: Chang Zou, Evelyn ...