Life in Shanghai 2025 Winter

2 minute read

Published:

I spent January and February 2026 interning at the LUMIA Lab at Shanghai Jiao Tong University. It was a busy time focused on ICML submissions, where I finished a first-author paper on dLLM and a third-author paper on PonderLM, among others. I chatted a lot with the seniors and my mentor in the group and learned a great deal.

Shanghai 2025 Winter
1) 3&1/2 Café with xvxv—I had the scallop drink and she had the wasabi cucumber; 2) I lost count of how many times we visited Hama Sushi; 3) German car restaurant where the pig was being roasted right next to us.

Outside the lab, I visited the Shanghai Institute of Innovation and toured around the city. I had a wonderful time with xvxv❤. We went café hopping, enjoyed a lot of Japanese food, and explored different cuisines. We even checked out the filming locations for The Good Days and B for Busy—seeing (exciting Lao Bai’s house).

Shanghai with xvxv
On my first day in Shanghai, the flowers, strawberries, and roast lamb from xvxv saved a bad day; We went to a beef buffet—binging on grilled durian and Häagen-Dazs—went shopping, visited the Bund, drank Algebraist, and had DQ; xvxv brought a bunch of aliens (e.g. LingHongWen); We used water to leave proof of our love on the window.

While Shanghai’s diverse and bustling city life is attractive, what really hooked me was the incredible research atmosphere and the powerful compute resources. To me, those are the real essentials for changing the world.

I wanna say thanks to xvxv. After some days it will be the Valentine’s Day. During these days, I wake up every morning to find you sleeping beside me. We get up, have a simple meal, and then might head out or just stay indoors. I feel a simplicity that is filled with happiness.

When I stay up late working on my drafts, you give me endless encouragement and hugs, taking care of me and keeping me company. In the future, we may face various challenges, but I believe everything we encounter will guide us in the right direction. We are like two trees growing side by side: our roots are tightly intertwined underground, and our leaves greet each other in the clouds.

You always give me the courage and motivation to push beyond my limits. We will be even closer than Gu and Jia’er in luschuanshuo, and I feel so, so lucky to have you in my life.

As Tagore wrote in Stray Birds:

“The world puts off its mask of vastness to its lover. It becomes small as one song, as one kiss of the eternal.”

Also, we decide to name our future dog “LES(OR RICE OR LICE) dog”, because of “lets go”. That is a pretty funny joke.

You May Also Enjoy

Future Blog Post

less than 1 minute read

Published:

This post will show up by default. To disable scheduling of future posts, edit config.yml and set future: false.

Thought

less than 1 minute read

Published:

I believe pretraining for large language models is far from finished. A lot of work today focuses on building agents and on post-training. Those steps can help on specific tasks, but the gains are often narrow. The deeper abilities we still need—making sense of the world and dealing with messy real situations—depend on large-scale pretraining with careful data and real changes to how the model is built. That includes new versions of the basic Transformer design, memory inside the model that ties ideas together more cleanly, and flexible ways to blend representations when the setting shifts.

Doing the obvious things right is not easy

less than 1 minute read

Published:

Doing the obvious things right is harder than it sounds. I have watched “model companies” that barely build models. I have seen placeholder papers with no real experiments—work you cannot reproduce, or results that do not add up. Plenty of people train models by hoping for a lucky run. And I am sure many of us have seen bad ideas pushed forward anyway, because someone cared more about their own gain than about whether the thing actually works.

Internship in Beijing as A Quant in 2025 Summer

5 minute read

Published:

I applied for countless internships in Shanghai, but somehow, I ended up in Beijing. I still remember the excitement when I received the internship offer in early July. I thought I was blazing a new trail, so I came here, rented an apartment, and started my internship life.