Principled Synthetic Data Enables the First Scaling Laws for LLMs in Recommendation Paper • 2602.07298 • Published 10 days ago • 2
Reasoning Cache: Continual Improvement Over Long Horizons via Short-Horizon RL Paper • 2602.03773 • Published 13 days ago • 9 • 3
Defeating the Training-Inference Mismatch via FP16 Paper • 2510.26788 • Published Oct 30, 2025 • 31 • 2