Scaling Domain Data Repetition in LLM Pretraining
Paper • 2608.14071 • Published • 17
LLM Training
-
-
Beyond the Black Box: A Survey on the Theory and Mechanism of Large Language Models
Paper • 2601.02907 • Published -
Large Language Model Post-Training: A Unified View of Off-Policy and On-Policy Learning
Paper • 2604.07941 • Published -
Data Science and Technology Towards AGI Part I: Tiered Data Management
Paper • 2602.09003 • Published • 12 -
Holistic Data Scheduler for LLM Pre-training via Multi-Objective Reinforcement Learning
Paper • 2606.24133 • Published • 12 -
Negative Self-Distillation: Learning to Reason by Avoiding Flaws
Paper • 2609.11699 • Published • 35 -
Releasing the Inner Jev: A Simple Recipe for Open Decision Models
Paper • 2610.17205 • Published • 1 -
Self-supervised Pretraining for Decision Foundation Model: Formulation, Pipeline and Challenges
Paper • 2401.00031 • Published -
LLM-as-Jev: LLMs Are Already Jev-Style Decision Models -- When and How to Fine-Tune Them
Paper • 2610.02076 • Published • 15 -
Can Jev be Your Q or Policy in Reinforcement Learning?
Paper • 2610.11692 • Published