Today I'm heading to San Diego for NeurIPS for the first time to present some research I've been working on. I got into machine learning about 10 years ago and always thought of NeurIPS as our field's Woodstock or Burning Man—the big one! I can't wait be in the mix! 1/N
Heading to neurips soon!
I'm on the job market and I work on making AI reliable, with a focus on healthcare.
Would love to meet people and hear about opportunities!
New paper alert 🚨
What if I told you there is an architecture that provides a _knob_ to control quality-efficiency trade-offs directly at test-time?
Introducing Compress & Attend Transformers (CATs) that provide you exactly this!
🧵(1/n) 👇
Many CDS faculty fellows have gone on to take tenure-track positions at top universities! This is one case where correlation definitely equals causation: https://t.co/E8el2hmdrK
Fellows work across all areas of research at CDS. I've personally enjoyed collaborating with past and current fellows in natural language and cognitive science (@najoungkim, @byungdoh, @ravfogel, and others!).
Apply to be a Faculty Fellow at the NYU Center for Data Science (CDS)! This is by default a two year position, with a light teaching load (one course a year), and full research independence, and opportunities to collaborate with CDS faculty and students. Deadline: November 24. 🧵
Fisher meets Feynman! 🤝
We use score matching and a trick from quantum field theory to make a product-of-experts family both expressive and efficient for variational inference.
To appear as a spotlight @ NeurIPS 2025.
#NeurIPS2025 (link below)
✨Masked Diffusion Language Models✨ are great for reasoning, but not just for the reasons you think!
Fast parallel decoding? 🤔 Any-order decoding? 🤨
Plot twist: MDLMs offer A LOT MORE for inference and post-training! 🎢🧵
What to do when you have zero rewards during RL?
We benchmarked RL baselines on a simple star-graph task where they underperform in zero reward scenarios.
Turns out, a dead simple data-centric intervention of just adding easy samples of the task helps unlock RL training! 👇
Does your company have a data science project to complete?
📢 Partner with NYU MS Data Science students this Fall!
📅 Submit your Capstone proposal by Aug 3: https://t.co/NrUN8lFIP3
Does your company have a data science project to complete?
📢 Partner with NYU MS Data Science students this Fall!
📅 Submit your Capstone proposal by Aug 3: https://t.co/NrUN8lFIP3
We will be presenting our work on steering and scaling diffusion models at #ICML2025!
Bonus:
- FK steering can beat grad guidance?! 🤯
- Boltz 🧬adapted FK steering to BioML!!! https://t.co/DhOqOfbi27
East Ex. Hall A-B #E-1308
(Tue 15 Jul 11 a.m. PDT — 1:30 p.m. PDT)
We are excited to announce a workshop on Demographic Diversity in Computer Vision (DemoDiv) at #CVPR2025!
Submit your work studying various axes of demographic diversity and fairness in models and datasets and join us in Nashville in June!
Deadline: March 31st
Got a diffusion model?
What if there were a way to:
- Get SOTA text-to-image prompt fidelity, with no extra training!
- Steer continuous and discrete (e.g. text) diffusions
- Beat larger models using less compute
- Outperform fine-tuning
- And keep your stats friends happy !?
What prompt generated the image on the right?
Come find out today at our tutorial on OOD generalization: Shortcuts, Spuriousness, and Stability
@Maggiemakar @aahladpuli
Panel: @ElanRosenfeld@AdtRaghunathan Danica Sutherland
Excited to announce that the Spurious Correlation and Shortcut Learning (SCSL) workshop has been accepted to #ICLR2025! 🎉 Join us to explore cutting-edge research on these critical topics.
Follow @SCSLWorkshop for updates and details!😉 https://t.co/UjZ6Zv1Wy1