Studying plant vegetative reproduction is key to increasing crop yield and for bioengineering. Kobe University research is making progress on studying the genetic regulation of the process in liverworts, which are ideal model plants and even a candidate for space crops.

Since the beginning of Earth's history, tiny particles of rock and metal from space have been hitting our planet. On clear nights, we can even see their traces as shooting stars. Trapped in layers of rock, these micrometeorites can remain preserved for billions of years. An international research team led by the University of Göttingen and including the Open University, the University of Pisa, and Leibniz University Hannover has developed a method that allows them to reconstruct the atmosphere of the past using fossilized micrometeorites. The results were published in Communications Earth & Environment.

A new paper led by a York University professor and published today in Nature Astronomy introduces a simple theoretical framework to describe the evolution of the coupled interior–atmosphere system of hot rocky exoplanets known as “lava planets.” The study combines expertise in geophysical fluid mechanics, exoplanetary atmospheres, and mineralogy to explore how the compositions of lava planets evolve through a process akin to distillation. What began as an exploratory study has since opened a promising new line of research. The predictions outlined in this work helped secure 100 hours of observation time on the James Webb Space Telescope. 

The new dataset introduces a seamless, annual Leaf-On Landsat composite data cube for China, spanning from 1985 to 2023.

Diffusion probabilistic models (DPMs) have achieved impressive success in high-resolution image synthesis, especially in recent large-scale text-to-image generation applications. An essential technique for improving the sample quality of DPMs is guided sampling, which usually needs a large guidance scale to obtain the best sample quality. The commonly-used fast sampler for guided

sampling is denoising diffusion implicit models (DDIM), a first-order diffusion ordinary differential equation (ODE) solver that generally needs 100 to 250 steps for high-quality samples. Although recent works propose dedicated high-order solvers and achieve a further speedup for sampling without guidance, their effectiveness for guided sampling has not been well-tested before. In this work, researchers demonstrate that previous high-order fast samplers suffer from instability issues, and they even become slower than DDIM when the guidance scale grows larger. To further speed up guided sampling, researchers propose DPM-Solver++, a high-order solver for the guided sampling of DPMs. DPM-Solver++ solves the diffusion ODE with the data prediction model and adopts thresholding methods to keep the solution matches training data distribution. Researchers further propose a multistep variant of DPM-Solver++ to address the instability issue by reducing the effective step size. Experiments show that DPM-Solver++ can generate high-quality samples within only 15 to 20 steps for guided sampling by pixel-space and latent-space DPMs.