Aahlad Manas Puli
Assistant Professor / Faculty Fellow at the Center for Data Science, NYU.
Email: aahlad [at] nyu [dot] edu
Publications
Here’s my Google Scholar profile. * denotes equal contribution.
2026
Attention and Compression is all you need for Controllably Efficient Language Models
Jatin Prakash, Aahlad Puli, Rajesh Ranganath.
In submission, 2026. [arXiv]
Flow Map Learning Via Non-Gradient Vector Flow
Mark Goldstein, Anshuk Uppal, Raghav Singhal, Aahlad Puli, Rajesh Ranganath.
ICLR, 2026. [OpenReview]
Extracting Representations in LLMs Robust to Distribution Shifts
Sweta Karlekar, Claudia Shi, Aahlad Puli, Carolina Zheng, Maggie Makar, Michal Kucer, John Bowlan, David Blei.
ICLR Workshop, 2026. [OpenReview]
2025
Black Box Causal Inference: Effect Estimation via Meta Prediction
Lucius E.J. Bynum*, Aahlad Puli*, Diego Herrero-Quevedo, Nhi Nguyen, Carlos Fernandez-Granda, Kyunghyun Cho, Rajesh Ranganath.
arXiv, 2025. [arXiv]
Let the Experts Speak: Improving Survival Prediction & Calibration via Mixture-of-Experts Heads
Todd Morrill, Aahlad Puli, Murad Megjhani, Soojin Park, Richard Zemel.
ML4H, 2025. [arXiv]
New-Onset Diabetes Assessment Using Artificial Intelligence-Enhanced Electrocardiography
Hao Zhang, Neil Jethani, Aahlad Puli, Leonid Garber, Lior Jankelson, Yindalon Aphinyanaphongs, Rajesh Ranganath.
ML4H, 2025. [arXiv]
Learning Is Not A Race: Improving Retrieval in Language Models via Equal Learning
Wanqian Yang, Aahlad Puli, Rajesh Ranganath.
EMNLP Findings, 2025. [ACL Anthology]
2024
Explanations that reveal all through the definition of encoding
Aahlad Puli*, Nhi Nguyen*, Rajesh Ranganath.
NeurIPS, 2024. [arXiv]
Contrasting with Symile: Simple Model-Agnostic Representation Learning for Unlimited Modalities
Adriel Saporta, Aahlad Puli, Mark Goldstein, Rajesh Ranganath.
NeurIPS, 2024. [arXiv]
Development and external validation of a dynamic risk score for early prediction of cardiogenic shock in cardiac intensive care units using machine learning
Yuxuan Hu, Albert Lui, Mark Goldstein, Mukund Sudarshan, Andrea Tinsay, Cindy Tsui, Samuel D Maidman, John Medamana, Neil Jethani, Aahlad Puli, Vuthy Nguy, Yindalon Aphinyanaphongs, Nicholas Kiefer, Nathaniel R Smilowitz, James Horowitz, Tania Ahuja, Glenn I Fishman, Judith Hochman, Stuart Katz, Samuel Bernard, Rajesh Ranganath.
European Heart Journal: Acute Cardiovascular Care, 2024. [journal]
Robust Anomaly Detection for Particle Physics Using Multi-Background Representation Learning
Abhijith Gandrakota, Lily Zhang, Aahlad Puli, Kyle Cranmer, Jennifer Ngadiuba, Rajesh Ranganath, Nhan Tran.
Machine Learning: Science and Technology, 2024. [journal]
Nuisances via Negativa: Adjusting for Spurious Correlations via Data Augmentation
Aahlad Puli, Nitish Joshi, Yoav Wald, He He, Rajesh Ranganath.
TMLR, 2024. [arXiv]
2023
Don’t blame Dataset Shift! Shortcut Learning due to Gradients and Cross Entropy
Aahlad Puli, Lily Zhang, Yoav Wald, Rajesh Ranganath.
NeurIPS, 2023. [arXiv]
Beyond Distribution Shift: Spurious Features Through the Lens of Training Dynamics
Nihal Murali, Aahlad Puli, Ke Yu, Rajesh Ranganath, Kayhan Batmanghelich.
TMLR, 2023. [arXiv]
New-Onset Diabetes Assessment Using Artificial Intelligence-Enhanced Electrocardiography
Lior Jankelson, Neil Jethani, Aahlad Puli, Hao Zhang, Leonid Garber, Yindalon Aphinyanaphongs, Rajesh Ranganath.
Heart Rhythm, 2023.
When More is Less: Incorporating Additional Datasets Can Hurt Performance By Introducing Spurious Correlations
Rhys Compton, Lily Zhang, Aahlad Puli, Rajesh Ranganath.
MLHC, 2023. [arXiv]
DIET: Conditional independence testing with marginal dependence measures of residual information
Mukund Sudarshan*, Aahlad Puli*, Wesley Tansey, Rajesh Ranganath.
AISTATS, 2023. [arXiv]
2022
OOD Generalization in the Presence of Nuisance-Induced Spurious Correlations
Aahlad Puli, Lily Zhang, Eric Oermann, Rajesh Ranganath.
ICLR, 2022. [arXiv]
Learning invariant representations with missing data
Mark Goldstein, Jörn-Henrik Jacobsen, Olina Chau, Adriel Saporta, Aahlad Puli, Rajesh Ranganath, Andrew Miller.
CLeaR, 2022. [arXiv]
Individual treatment effect estimation in the presence of unobserved confounding using proxies: a cohort study in stage III non-small cell lung cancer
Wouter AC van Amsterdam, Joost JC Verhoeff, Netanja I Harlianto, Gijs A Bartholomeus, Aahlad Puli, Pim A de Jong, Tim Leiner, Anne SR van Lindert, Marinus JC Eijkemans, Rajesh Ranganath.
Nature Scientific Reports, 2022. [journal]
2021 and earlier
Inverse-Weighted Survival Games
Xintian Han, Mark Goldstein, Aahlad Puli, Thomas Wies, Adler J Perotte, Rajesh Ranganath.
NeurIPS, 2021. [proceedings]
CONTRA: Contrarian statistics for controlled variable selection
Mukund Sudarshan, Aahlad Puli, Lakshmi Subramanian, Sriram Sankararaman, Rajesh Ranganath.
AISTATS, 2021. [proceedings]
Causal Estimation with Functional Confounders
Aahlad Puli, Adler J Perotte, Rajesh Ranganath.
NeurIPS, 2020. [proceedings]
General Control Functions for Causal Effect Estimation from IVs
Aahlad Puli, Rajesh Ranganath.
NeurIPS, 2020. [proceedings]
X-CAL: Explicit calibration for survival analysis
Mark Goldstein*, Xintian Han*, Aahlad Puli*, Adler J Perotte, Rajesh Ranganath.
NeurIPS, 2020. [proceedings]
Removing hidden confounding by experimental grounding
Nathan Kallus, Aahlad Puli, Uri Shalit.
NeurIPS, 2018. [proceedings]