Yu Changqian

Changqian

·

https://yu-changqian.github.io/

AI & ML interests

Current: Multi-modal LLM / LLM / Diffusion Model Previous: Autonomous Driving (including: perception/prediction) / 2D&3D Detection&Segmentation

Recent Activity

upvoted a paper 11 days ago

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

submitted a paper 11 days ago

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

published a dataset about 1 month ago

Kwai-Kolors/LongBanana

View all activity

Organizations

upvoted a paper 11 days ago

StableVQ: Practical Guidelines for Stable Vector-Quantized Tokenizer Training

Paper • 2609.26774 • Published 13 days ago • 56

upvoted a paper about 2 months ago

AdvFD: Boosting Visual Generation via Adversarial Fr'echet Distance Loss

Paper • 2608.11205 • Published Aug 11 • 27

upvoted a paper 3 months ago

MetaView: Monocular Novel View Synthesis with Scale-Aware Implicit Geometry Priors

Paper • 2607.12000 • Published Jul 13 • 32

upvoted a paper 11 months ago

Sample By Step, Optimize By Chunk: Chunk-Level GRPO For Text-to-Image Generation

Paper • 2510.21583 • Published Oct 24, 2025 • 31

upvoted 2 papers over 1 year ago

Kwai Keye-VL Technical Report

Paper • 2507.01949 • Published Jul 2, 2025 • 133

Ingredients: Blending Custom Photos with Video Diffusion Transformers

Paper • 2501.01790 • Published Jan 3, 2025 • 8

upvoted a paper about 2 years ago

FLUX that Plays Music

Paper • 2409.00587 • Published Sep 1, 2024 • 33

upvoted a paper over 2 years ago

Diffusion-RWKV: Scaling RWKV-Like Architectures for Diffusion Models

Paper • 2404.04478 • Published Apr 6, 2024 • 13