Qingyan PRO

QingyanBai

·

https://bqy.info/

AI & ML interests

Multimodal Generative Models and Agents.

Recent Activity

upvoted a paper 1 day ago

WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory

upvoted a paper 6 days ago

SMELT: Scaling Laws for Compute-Matched MoE Looped Transformers

upvoted a paper 6 days ago

Dream-RSI: Recursive Self-Improvement through Evolving Worlds

View all activity

Organizations

Papers 5

arxiv:2607.07534

arxiv:2601.20540

arxiv:2510.15742

arxiv:2412.21079

spaces 1

Ditto

Generate edited videos based on text instructions

models 1

QingyanBai/Ditto_models

Video-to-Video • Updated Jun 21 • 76

datasets 1

QingyanBai/Ditto-1M

Updated Jun 1 • 5.75k • 52