Kangfu Mei

Research Scientist at @GoogleDeepMind. | Ph.D. @JohnsHopkins. | Tweets are my own.

[Kangfu Mei @KangfuM]

🚀🚀🚀

[Arena.ai @arena]

Exciting news: Gemini Omni Flash is now #1 in the Video Arena (both Text-to-Video and Image-to-Video)!

For Text-to-Video this is a massive +158 pt improvement over Veo 3.1 (1080p) and a large +61 pt lead over the next best model, Seedance 2.0.

Congrats @GoogleDeepMind for this huge milestone!

[Kangfu Mei @KangfuM]

I will be in Denver for CVPR from June 3rd to 6th! 🏔️

I'm looking forward to connecting and discussing anything related to world models and verifiable rewards.

Or just even stress for graduation and job hunting, unhappiness of paper rejection, are all welcome.

[Ethan Mollick @emollick]

I think people don't realize why Gemini Omni is different than other video AIs. It is fully multimodal, so it can edit video natively, too

I took the famous "train " movie from 1896 & made it a bullet train, LEGO, added a time traveler, a centipede, muppets... (see reflections?)

[Kangfu Mei @KangfuM]

@vishalm_patel congratulations!

[Kangfu Mei @KangfuM]

Let's rock! 🎸🎸🎸

[Sundar Pichai @sundarpichai]

On our way to I/O 2026. See you at 10am PT tomorrow!

[Kangfu Mei @KangfuM]

@CeyuanY Congratulations Ceyuan! That’s great work!

[Kangfu Mei @KangfuM]

Congratulations Yuyang!

[Computational Imaging Group (CIG) @uwcig]

Congratulations to Dr. Yuyang Hu for successfully defending his PhD at WashU ESE. We are very impressed with all the research you have done over the past few years and wishing you even greater achievements in the years to come! 🎉

[Kangfu Mei @KangfuM]

@EvelynZ5699647 This is a great work. Any plan of applying this into video generative model as RL rewards?

[Logan Kilpatrick @OfficialLoganK]

Video’s here to stay - introducing Veo 3.1 Lite, our most cost efficient video generation model to date, and on April 7th we are also reducing the price for Veo 3.1 Fast : )

[Demis Hassabis @demishassabis]

Ten years ago, AlphaGo’s legendary match in Seoul heralded the start of the modern era in AI. Its famous ‘Move 37’ signaled to us that AI techniques were ready to tackle real-world problems in areas like science - and ideas inspired by these methods are critical to building AGI

[Kangfu Mei @KangfuM]

@baaadas @LumaLabsAI Great job 👏 congratulations

[Kangfu Mei @KangfuM]

Congratulations to the team! It is just the first month of 2026.

[Google DeepMind @GoogleDeepMind]

Step inside Project Genie: our experimental research prototype that lets you create, edit, and explore virtual worlds. 🌎

[Kangfu Mei @KangfuM]

🧞‍♂️

[Kangfu Mei @KangfuM]

Happy to share that "Streaming Autoregressive Video Generation via Diagonal Distillation" is accepted by #ICLR26 ! See you all in Brazil 🇧🇷

[Kangfu Mei @KangfuM]

@CuiMao 这效果也太好了🤣

[Kangfu Mei @KangfuM]

Everyone knows pretraining is powerful, but its potential in IR tasks beyond SFT or RL remains largely unexplored.

In this work, we show that reusing pretraining (e.g., Flux) is easier than you think!

🧱 LEGO doesn't need pair-wise data and converts pretrained knowledge in an unsupervised (pseudo-supervised) way.

Huge congrats to our intern @YuyangHu_666 for this great work! This is my final paper of 2025. Can't wait to show you what we have for an even more exciting 2026! 🚀

[Peyman Milanfar @docmilanfar]

LEGO: our post-training framework adapts diffusion models to unseen domains without paired ground truth. Using large-scale image generation models as reference we synthesize "oracle" training pairs

LEGO creates sharp high quality results on real images where others fail

1/5

[Peyman Milanfar @docmilanfar]

Five generations of our team's interns, all doing so great in their careers. Proud of @_vztu , @KangfuM , @YuyangHu_666 , @hyungjin_chung , and @giannis_daras

[Kangfu Mei @KangfuM]

Big congratulations 🎉🎊🍾

[Vishal Patel @vishalm_patel]

Honored to be named an IEEE Fellow for contributions to image processing, computer vision & biometrics. Also grateful to be an AAAI Senior Member and a 2025 Clarivate Highly Cited Researcher. Huge thanks to my students, mentors & collaborators! @jhuclsp @HopkinsEngineer

[Kangfu Mei @KangfuM]

Great work Yiqun!

[Yiqun Mei @myq_1997]

Thanks @_akhaliq for sharing our work.
Visit our webpage: https://t.co/dT1vZ8bSqG to learn more.

[Google DeepMind @GoogleDeepMind]

Hello from #NeurIPS2025 👋

We're hosting a range of sessions at the @Google booth, including a Q&A with @JeffDean and the Gemini team, plus demos like SIMA 2 – our AI agent for 3D virtual worlds.

See the full schedule → https://t.co/B1TQ335HoS