@majhas.bsky.social
PhD Student at Mila & University of Montreal | Generative modeling, sampling, molecules
majhas.github.io
majhas.github.io
(1/n)🚨Train a model solving DFT for any geometry with almost no training data
Introducing Self-Refining Training for Amortized DFT: a variational method that predicts ground-state solutions across geometries and generates its own training data!
📜 arxiv.org/abs/2506.01225
💻 github.com/majhas/self-...
Introducing Self-Refining Training for Amortized DFT: a variational method that predicts ground-state solutions across geometries and generates its own training data!
📜 arxiv.org/abs/2506.01225
💻 github.com/majhas/self-...
June 10, 2025 at 7:49 PM
(1/n)🚨Train a model solving DFT for any geometry with almost no training data
Introducing Self-Refining Training for Amortized DFT: a variational method that predicts ground-state solutions across geometries and generates its own training data!
📜 arxiv.org/abs/2506.01225
💻 github.com/majhas/self-...
Introducing Self-Refining Training for Amortized DFT: a variational method that predicts ground-state solutions across geometries and generates its own training data!
📜 arxiv.org/abs/2506.01225
💻 github.com/majhas/self-...
Reposted
New preprint! 🧠🤖
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
June 6, 2025 at 5:40 PM
New preprint! 🧠🤖
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
How do we build neural decoders that are:
⚡️ fast enough for real-time use
🎯 accurate across diverse tasks
🌍 generalizable to new sessions, subjects, and even species?
We present POSSM, a hybrid SSM architecture that optimizes for all three of these axes!
🧵1/7
Reposted
🧵(1/7) Have you ever wanted to combine different pre-trained diffusion models but don't have time or data to retrain a new, bigger model?
🚀 Introducing SuperDiff 🦹♀️ – a principled method for efficiently combining multiple pre-trained diffusion models solely during inference!
🚀 Introducing SuperDiff 🦹♀️ – a principled method for efficiently combining multiple pre-trained diffusion models solely during inference!
December 28, 2024 at 2:32 PM
🧵(1/7) Have you ever wanted to combine different pre-trained diffusion models but don't have time or data to retrain a new, bigger model?
🚀 Introducing SuperDiff 🦹♀️ – a principled method for efficiently combining multiple pre-trained diffusion models solely during inference!
🚀 Introducing SuperDiff 🦹♀️ – a principled method for efficiently combining multiple pre-trained diffusion models solely during inference!
Reposted
🔊 Super excited to announce the first ever Frontiers of Probabilistic Inference: Learning meets Sampling workshop at #ICLR2025 @iclr-conf.bsky.social!
🔗 website: sites.google.com/view/fpiwork...
🔥 Call for papers: sites.google.com/view/fpiwork...
more details in thread below👇 🧵
🔗 website: sites.google.com/view/fpiwork...
🔥 Call for papers: sites.google.com/view/fpiwork...
more details in thread below👇 🧵
December 18, 2024 at 7:09 PM
🔊 Super excited to announce the first ever Frontiers of Probabilistic Inference: Learning meets Sampling workshop at #ICLR2025 @iclr-conf.bsky.social!
🔗 website: sites.google.com/view/fpiwork...
🔥 Call for papers: sites.google.com/view/fpiwork...
more details in thread below👇 🧵
🔗 website: sites.google.com/view/fpiwork...
🔥 Call for papers: sites.google.com/view/fpiwork...
more details in thread below👇 🧵
Reposted
Now you can generate equilibrium conformations for your small molecule in 3 lines of code with ET-Flow! Awesome effort put in by @fntwin.bsky.social!
December 12, 2024 at 4:37 PM
Now you can generate equilibrium conformations for your small molecule in 3 lines of code with ET-Flow! Awesome effort put in by @fntwin.bsky.social!
Reposted
ET-Flow shows, once again, that equivariance is better than Transformer when physical precision matters!
come see us at @neuripsconf.bsky.social !!
come see us at @neuripsconf.bsky.social !!
December 7, 2024 at 3:57 PM
ET-Flow shows, once again, that equivariance is better than Transformer when physical precision matters!
come see us at @neuripsconf.bsky.social !!
come see us at @neuripsconf.bsky.social !!
Excited to share our work! I had a wonderful time collaborating with these brilliant people
We’re excited to present ET-Flow at #NeurIPS 2024—an Equivariant Flow Matching model that combines simplicity, efficiency, and precision to set a new standard for 3D molecular conformer generation.
🔖Paper: arxiv.org/abs/2410.22388
🔗Github: github.com/shenoynikhil...
🔖Paper: arxiv.org/abs/2410.22388
🔗Github: github.com/shenoynikhil...
December 7, 2024 at 4:01 PM
Excited to share our work! I had a wonderful time collaborating with these brilliant people