Tomer Galanti

On the Alignment Between Supervised and Self-Supervised Contrastive Learning

Scalable Principal–Agent Contract Design via Gradient-Based Optimization

LLM-ERM: Sample-Efficient Program Learning via LLM-Guided Search

DisCO: Reinforcing Large Reasoning Models with Discriminative Constrained Optimization

Formation of Representations in Neural Networks

Self-Supervised Contrastive Learning is Approximately Supervised Contrastive Learning

Distributed Speculative Inference (DSI): Speculation Parallelism for Provably Faster Lossless Language Model Inference

The Fair Language Model Paradox

Type-II Saddles and Probabilistic Stability of Stochastic Gradient Descent