Abstract: This paper theoretically applied the total channel number of Braginskii switch behavior on the cloud-to-ground (CG) flash multiplicity estimation with the coaxial unification fractal model ...

This repository provides the official implementation for the paper "Contrastive Mean Teacher for Robust Low-Light Image Enhancement", Arxiv, 2025. [2025-06-28] The test codes, pre-trained weights and ...

VLM-3R is a unified Vision-Language Model (VLM) framework integrating 3D reconstructive instruction tuning for deep spatial understanding from monocular video. The rapid advancement of Large ...