ICML 2026 Presentations from our group

Main meeting

1. A Random Matrix Perspective on the Consistency of Diffusion Models
Binxu Wang ⋅ Jacob A Zavatone-Veth ⋅ Cengiz Pehlevan
Oral Thu, Jul 9, 2026 • 3:00 AM – 3:15 AM EDT ASEM BALLROOM 201-203
Poster Thu, Jul 9, 2026 • 4:00 AM – 5:45 AM EDT

2. Demystifying LLM-as-a-Judge: Analytically Tractable Model for Inference-Time Scaling
Indranil Halder ⋅ Cengiz Pehlevan
Poster Tue, Jul 7, 2026 • 9:30 PM – 11:15 PM EDT HALL A

3. CompleteP for RL: Maintaining Feature Learning When Scaling Deep Reinforcement Learning
M Ganesh Kumar ⋅ Adam Lee ⋅ Blake Bordelon ⋅ Cengiz Pehlevan
PosterTue, Jul 7, 2026 • 1:00 AM – 2:45 AM EDT HALL A

4. Universal One-third Time Scaling in Learning Peaked Distributions
Yizhou Liu ⋅ Ziming Liu ⋅ Cengiz Pehlevan ⋅ Jeff Gore
PosterTue, Jul 7, 2026 • 1:00 AM – 2:45 AM EDT

5. Hyperparameter Transfer with Mixture-of-Expert Layers
Tianze Jiang ⋅ Blake Bordelon ⋅ Cengiz Pehlevan ⋅ Boris Hanin
Poster Thu, Jul 9, 2026 • 4:00 AM – 5:45 AM EDT HALL A

Workshops

1. Practical Bayesian Optimization for Scientific Discovery
Hamza Chaudhry ⋅ Sean Murphy ⋅ Umesh Padia ⋅ Cengiz Pehlevan ⋅ James Harrison ⋅ George Church ⋅ Jasper Snoek
Decision-Making from Offline Datasets to Online Adaptation: Black-Box Optimization to Reinforcement Learning
AI for Science: AI Scientists -- Tools, Co-authors, or Founders?
 

2. Anytime Pretraining: Horizon-Free Learning-Rate Schedules with Weight Averaging
Alex Meterez ⋅ Pranav Nair ⋅ Depen Morwani ⋅ Cengiz Pehlevan ⋅ Sham Kakade
High Dimensional Learning Dynamics: the Science of Scaling

3. Sequential Correlations Change In-Context Learning: Effective Context Length and Architectural Mismatch
Mary Letey ⋅ Yue Lu ⋅ Cengiz Pehlevan ⋅ Jacob A Zavatone-Veth
High Dimensional Learning Dynamics: the Science of Scaling

4. Invited talk
Cengiz Pehlevan
Foundations of Deep Generative Models: Understanding Memorization, Generalization, and Reasoning