Mining potential of open-source VLMs! Mini-Gemini is a novel framework ranges from 2B to 34B VLMs for hi-resolution image understanding. It has a impressive OCR capability, and can generate HQ images powered by its multi-modal reasoning ability.

The slow agent decomposes the task and determines "which actions" to learn. The fast agent writes code and RL configurations for low-level execution.

Rethinking the inversion process. Boosting Diffusion-based Editing with 3 Lines of Code.

Add 'Lego' attribute to the child, an edited video is generated. Powered by a novel video inversion process and cross-attention control. We also find that a Decoupled-Guidance strategy is essential for video editing.

A simple transformer is already able to synthesize high-quality images.

Target accuracy is the goal, so we argue for optimizing as much as possible on the target data.

Tent equips a model to adapt itself to new and different data during testing.

Learn visual representation in hyperbolic space for Zero-Shot Recognition.

Morph Studio, which has its own text-to-video model, just introduced an AI filmmaking platform. The eponymous tool takes the form of a storyboard, where users can create and edit shots by entering text prompts for different scenes and combine them into a cohesive narrative. It has partnership with Stability AI.