Vahab Mirrokni
- Efficient Data Selection at Scale via Influence Distillation
- Nested Learning: The Illusion of Deep Learning Architectures
- PiKE: Adaptive Data Mixing for Large-Scale Multi-Task Learning Under Low Gradient Conflicts
- Self-Boost via Optimal Retraining: An Analysis via Approximate Message Passing
- Titans: Learning to Memorize at Test Time