Julia Kempe
Silver Professor of Computer Science, Mathematics and Data Science
Studies the foundations of machine learning, including scaling behavior, synthetic data and robustness, alongside quantum computing.
Julia Kempe's lab at NYU works on trustworthy & safe ML. 30 papers since 2023, 26 at top venues, mostly ICLR and ICML. 4 PhD students. 10 alumni, most often to Meta. Applications close .
At a glance
- papers, 2023–2026
- 8/ year
- at T1 venues
- 87%
- PhD students
- 4
- MS students
- 0
- alumni listed
- 10
Papers by yearfrom the lab homepage, 2023–2026
- 9 papers2026 · T1 9 · collecting
- 7 papers2025 · T1 6
- 12 papers2024 · T1 10
- 2 papers2023 · T1 1
T1 venuesOtherVenues: ICLR, ICML, NeurIPS, TMLR, AISTATS
Research focuspaper titles that use each topic's words
- scaling2 papers
- Trustworthy & safe ML1 paper
- synthetic data0 papers
- quantum computing0 papers
- alongside0 papers
Paper list
From the lab homepage.
20269 papers
- ICML (T1)Efficient RL Training for LLMs with Experience Replay (opens in a new tab)
- ICML (T1)Embedding Trust: Semantic Isotropy Predicts Nonfactuality in Long-Form Text Generation (opens in a new tab)
- ICLR (T1)From Concepts to Components: Concept-Agnostic Attention Module Discovery in Transformers (opens in a new tab)
- ICLR (T1)How reinforcement learning after next-token prediction facilitates learning (opens in a new tab)
- TMLR (T1)On the Geometry of Regularization in Adversarial Training: High-Dimensional Asymptotics and Generalization Bounds (opens in a new tab)
- ICLR (T1)OpenApps: Simulating Environment Variations to Measure UI-Agent Reliability (opens in a new tab)
- ICLR (T1)Soft Tokens, Hard Truths (opens in a new tab)
- ICML (T1)Teaching Models to Teach Themselves: Reasoning at the Edge of Learnability (opens in a new tab)
- ICML (T1)What Characterizes Effective Reasoning? Revisiting Length, Review, and Structure of Chain-of-Thought (opens in a new tab)
20257 papers
- NeurIPS (T1)Asymmetric REINFORCE for Off-Policy Reinforcement Learning: Balancing Positive and Negative Rewards (opens in a new tab)
- ICLR (T1)Beyond Model Collapse: Scaling Up with Synthesized Data Requires Verification (opens in a new tab)
- ICLR (T1)DRoP: Distributionally Robust Data Pruning (opens in a new tab)
- ICLR (T1)Flavors of Margin: Implicit Bias of Steepest Descent in Homogeneous Neural Networks (opens in a new tab)
- NeurIPS WorkshopOutcome-based Exploration for LLM Reasoning (opens in a new tab)
- ICML (T1)PILAF: Optimal Human Preference Sampling for Reward Modeling (opens in a new tab)
- ICLR (T1)Strong Model Collapse (opens in a new tab)
202412 papers
- ICML (T1)A Tale of Tails: Model Collapse as a Change of Scaling Laws (opens in a new tab)
- TMLR (T1)Attacking Bayes: On the Adversarial Robustness of Bayesian Neural Networks (opens in a new tab)
- ICML (T1)Deconstructing the Goldilocks Zone of Neural Network Initialization (opens in a new tab)
- ICLR (T1)Embarrassingly Simple Dataset Distillation (opens in a new tab)
- NeurIPS WorkshopEmergent properties with repeated examples (opens in a new tab)
- NeurIPS (T1)Iteration Head: A Mechanistic Study of Chain-of-Thought (opens in a new tab)
- Journal of Statistical Mechanics: Theory and ExperimentKernels, Data & Physics (Lecture Notes, Les Houches 2022) (opens in a new tab)
- AISTATS (T1)Mind the GAP: Improving Robustness to Subpopulation Shifts with Group-Aware Priors (opens in a new tab)
- NeurIPS (T1)Mission Impossible: A Statistical Perspective on Jailbreaking LLMs (opens in a new tab)
- NeurIPS (T1)Model Collapse Demystified: The Case of Regression (opens in a new tab)
- TMLR (T1)On the Robustness of Neural Collapse and the Neural Collapse of Robustness (opens in a new tab)
- NeurIPS (T1)The Price of Implicit Bias in Adversarially Robust Generalization (opens in a new tab)
Where alumni went10 listed, 100% to industry
Industry 3Academia 0
Most often to
- Meta
Apply to NYUFall 2027
Computer Science PhD (Courant / Tandon)
- Round
- Single round (December)
- Opens
- —
- GRE
- See program page
- English
- See program page
- Fee
- See program page
Not yet confirmed for this intake.
Similar labsnearest by research summary
- CNTsinghua
Keyou You
- Reinforcement learning
- USCMU
Zachary C. Lipton
- Trustworthy & safe ML
- USUCSD
Sicun Gao
- Reinforcement learning
- USNYU
Christopher Musco
- Data management & big data
- USUMD
Heng Huang
- Learning theory & optimization
- USUCSD
Yoav Freund
- Learning theory & optimization
Your notes
Save this lab to keep notes.
Data
Checked against the lab homepage.
- Collected
- Verified
- Papers
- The lab homepage
- Members
- Lab homepage
- Alumni
- Lab homepage
Tiers are venue classifications, not lab rankings.
NYUcloses