@aahladpuli profile picture
aahlad puli @aahladpuli

On the 25-26 job market! Faculty Fellow at NYU. AI/ML, OOD generalization, healthcare, causality. scholar.google.com/citations…

Joined December 2010
  • Tweets 603
  • Following 1,589
  • Followers 579
39 Photos and videos
aahlad puli retweeted
Todd Morrill
@tee_oh_double_d
30 Nov 2025
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 Manβ€”the big one! I can't wait be in the mix! 1/N
2
1
1
251
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!
1
4
333
Simple, efficient, and surprisingly effective.
Jatin Prakash@bicycleman15
10 Nov 2025
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) πŸ‘‡
2
442
aahlad puli retweeted
Tal Linzen
@tallinzen
29 Oct 2025
It's also a pretty good deal financially! Here's the link to apply: apply.interfolio.com/174686
1
3
1,248
aahlad puli retweeted
Tal Linzen
@tallinzen
29 Oct 2025
Many CDS faculty fellows have gone on to take tenure-track positions at top universities! This is one case where correlation definitely equals causation: cds.nyu.edu/faculty-fellow-a…

1
1
2
1,415
aahlad puli retweeted
Tal Linzen
@tallinzen
29 Oct 2025
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!).
1
1
3
710
Join us!
Tal Linzen
@tallinzen
29 Oct 2025
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. 🧡
3
752
So cool
Diana Cai
@dianarycai
27 Oct 2025
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)
Fisher meets Feynman: score-based variational inference with a product of experts

ALT Fisher meets Feynman: score-based variational inference with a product of experts
4
432
aahlad puli retweeted
Zachary Horvitz
@zachary_horvitz
27 Oct 2025
✨Masked Diffusion Language Models✨ are great for reasoning, but not just for the reasons you think! Fast parallel decoding? πŸ€” Any-order decoding? 🀨 Plot twist: MDLMs offer A LOT MORE for inference and post-training! 🎒🧡
4
38
163
21,846
In Dyson's frogs/birds analogy, did he intend the subtext that the wells are all connected via groundwater?
1
134
AI peeps, what are your favorite AI for health papers from recent times?
1
1
3
304
aahlad puli retweeted
Jatin Prakash@bicycleman15
13 Sep 2025
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! πŸ‘‡
Anirudh Buvanesh@AnirudhBuvanesh
13 Sep 2025
Zero rewards after tons of RL training? 😞 Before using dense rewards or incentivizing exploration, try changing the data. Adding easier instances of the task can unlock RL training. πŸ”“πŸ“ˆTo know more checkout our blog post here: spiffy-airbus-472.notion.sit…. Keep reading 🧡(1/n)
2
7
15
875
Great opportunity to work with NYU CDS students!
NYU Center for Data Science
@NYUDataScience
25 Jun 2025
Does your company have a data science project to complete? πŸ“’ Partner with NYU MS Data Science students this Fall! πŸ“… Submit your Capstone proposal by Aug 3: docs.google.com/forms/d/e/1F…
1
5
700
aahlad puli retweeted
NYU Center for Data Science
@NYUDataScience
25 Jun 2025
Does your company have a data science project to complete? πŸ“’ Partner with NYU MS Data Science students this Fall! πŸ“… Submit your Capstone proposal by Aug 3: docs.google.com/forms/d/e/1F…
3
2
3,410
Easy, elegant, effective
Zachary Horvitz
@zachary_horvitz
12 Jul 2025
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!!! x.com/GabriCorso/status/1916… East Ex. Hall A-B #E-1308 (Tue 15 Jul 11 a.m. PDT β€” 1:30 p.m. PDT)
1
3
278
aahlad puli retweeted
Polina Kirichenko
@polkirichenko
21 Feb 2025
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
2
15
97
28,182
Simple and effective
Raghav Singhal@_rk_singhal
21 Jan 2025
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 !?
5
420
aahlad puli retweeted
Yoav Wald@wald_yoav
10 Dec 2024
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
5
19
5,012
aahlad puli retweeted
Hesam Asadollahzadeh@HesamAsdz
3 Dec 2024
Excited to announce that the Spurious Correlation and Shortcut Learning (SCSL) workshop has been accepted to #ICLR2025! πŸŽ‰ Join us to explore cutting-edge research on these critical topics. Follow @SCSLWorkshop for updates and details!πŸ˜‰ openreview.net/forum?id=B5Ml…

Luis Oala
@LuisOala
3 Dec 2024
2025 @iclr_conf workshops are now available at openreview.net/group?id=ICLR… thank you to all the amazing teams that prepared proposals out of 120 strong submissions 40 proposals were selected for 2 days of workshops in singapore a blog post with links to workshop sites will follow
7
21
1,904
Say your AI makes great predictions. You wonder which inputs power prediction. You run an algorithm which selects inputs that are said to predict at 100% accuracy. The selected input values seem to be implausibly predictive (see image); this confuses you. What do you do now?
1
3
12
833
more replies
@nhinn__ and I will be at Neurips, presenting this paper. Come chat with us at Poster Session 3 East. Thursday, 12 Dec; 11 a.m. PST β€” 2 p.m. PST
1
1
3
189
Of potential interest to @SudarshanMukund @sarahookr @hima_lakkaraju @apsarathchandar @zacharylipton @jen_hsia @danish037 @neiljethani
218
Load more