Tony Shen
I am a PhD student in Computing Science at Simon Fraser University, supervised by Dr. Martin Ester. My research focuses on structure-based drug design with generative flow networks and flow matching models.
Currently, I am also a visiting researcher at Northeastern University and Broad Institute of MIT and Harvard, where I work on foundation models for bio-molecular generation and antibiotics discovery applications under the supervision of Dr. Wengong Jin.
Previously, I worked as a Machine Learning Consultant at Transcripta Bio, where I conducted experiments on generative models and large scale virtual screening. Before that, I interned at Recursion, developing the map of Biology and Google, working on AI models for search.
Research Interests
My research is centered around AI for Drug Discovery, with a specific focus on:
- Foundation models for bio-molecular generation
- Generative Flow Networks (GFlowNets)
- Flow Matching and Diffusion Models
- Structure-based drug design
news
May 01, 2025 | Our paper “Compositional Flows for 3D Molecule and Synthesis Pathway Co-design” has been accepted at ICML 2025. See you in Vancouver 🇨🇦! |
---|---|
Mar 05, 2025 | Our paper “Compositional Flows for 3D Molecule and Synthesis Pathway Co-design” has been selected as oral presentation at GEM and AI4MAT workshops at ICLR 2025! |
Jan 22, 2025 | Our paper “Generative Flows on Synthetic Pathway for Drug Design” has been accepted at ICLR 2025 in Singapore 🇸🇬! |
Sep 03, 2024 | Our paper “TacoGFN: Target Conditioned GFlowNet for Structure-based Drug Design” has been accepted at TMLR! |
Oct 27, 2023 | Our paper “TacoGFN: Target Conditioned GFlowNet for Drug Design” has been accepted as a spotlight at the GenBio workshop and as a poster at the AI4D3 workshop at NeurIPS 2023! |
selected publications
- ICMLCompositional Flows for 3D Molecule and Synthesis Pathway Co-designInternational Conference on Machine Learning (ICML), 2025
- ICLRGenerative Flows on Synthetic Pathway for Drug DesignInternational Conference on Learning Representations (ICLR), 2025
- TMLRTacoGFN: Target Conditioned GFlowNet for Structure-Based Drug DesignTransactions on Machine Learning Research (TMLR), 2024