Stories by lnyan

Visualizing 6D Mesh Parallelism

FlowEdit: Inversion-Free Text-Based Editing Using Pre-Trained Flow Models

3DTrajMaster: Mastering 3D Trajectory for MultiEntity Motion in Video Generation

SynCamMaster: Synchronizing MultiCamera Video Generation from Diverse Viewpoints

txt.dev: a writing tool that deliberately stays simple

Flow Matching Guide and Code

Neural Attention Memory Model: An Evolved Universal Transformer Memory

Yazi: Fast terminal file manager written in Rust, based on async I/O

Trellis: Structured 3D Latents for Scalable and Versatile 3D Generation

Necomimi Illusion: Generating Ownership of Cat Ears w/ Haptic Feedback via Hair

Motion Prompting: Controlling Video Generation with Motion Trajectories

Tencent Hunyuan video: open-source text-to-video generation model

Diffusion Meets Flow Matching: Two Sides of the Same Coin

Exploring the Power of Parallelized CPU Architectures

Capacitive Touch Keyboard Business Card

AC3D: Analyzing and Improving 3D Camera Control in Video Diffusion Transformers

Quark: Real-Time, High-Resolution, and General Neural View Synthesis

Generative Omnimatte: Learning to Decompose Video into Layers

DeepSpeed-Domino: Communication-Free LLM Training Engine

Cat-human interaction from the viewpoint of physics: an equation of motion

1
2
3
4
5
6
7
8
9
10