Chen Lin

@bose_joey @imperialcollege @ICComputing Congrats Joey!💪

🎉Personal update: I'm thrilled to announce that I'm joining Imperial College London @imperialcollege as an Assistant Professor of Computing @ICComputing starting January 2026. My future lab and I will continue to work on building better Generative Models 🤖, the hardest AI4Science applications in computational biology 🧬and chemistry 🧪, and also a sprinkling of Deep Learning theory 📚 that supports these goals.

New work with @lars__schaaf, @WillLin1028, @wanggrun, and @philiptorr: We optimize neural networks to smoothly represent minimum energy paths and predict transition states for chemical reactions. Compared to the traditional approach, our method shows (i) improved resilience to the initial guess, (ii) easy adaptability to escape local minima, (iii) the ability to capture a complex path on its own, and (iv) potential to generalize to unseen systems. This offers a flexible alternative to discrete methods that could unlock building a universal reaction path predictor. Our paper should appeal to anyone interested in machine learning to advance computational chemistry. Preprint: arxiv.org/abs/2502.15843.

New preprint! 🚨 We scale equilibrium sampling to hexapeptide (in cartesian coordinates!) with Sequential Boltzmann generators!  📈 🤯 Work with @bose_joey, @WillLin1028, @leonklein26, @mmbronstein and @AlexanderTong7 Thread 🧵 1/11

@Francesco_dgv Looking forward to it! 💪💪💪

Really looking forward to giving this talk! I will provide an overview of my research on oversquashing and graph rewiring over and I will discuss future directions I am excited to work on within and beyond GNNs 🦕

Are RGB inputs good enough for open-world generalization of object detection? Excited to share our #ICLR2023 paper GOOD -- we show that geometric cues can significantly boost the performance for the task! arxiv.org/abs/2212.11720 (Joint work with @AutoVisionGroup @isDanZhang )

How can we leverage successful pretraining techniques from transformers to improve purely convolutional networks? The answer is *Sparse Convolutions*! Let's see what happens when purely convolutional networks are pretrained with 1.28 million unlabeled images ... 1/7

Designing BERT for convolutional networks: sparse and hierarchical masked modeling Keyu Tian, Yi Jiang, Qishuai Diao, Chen Lin, Liwei Wang, Zehuan Yuan tl;dr: create image, which looks to CNN same, as transformers -> MIM starts working arxiv.org/abs/2301.03580…

The paper designs a working MLM style pretraining for convnets! Huge if this really works.