← Back to Benchmarks
simmediumoffline-rlmetric · varies

Towards a Unified View of Large Language Model Post-Training

Description

Two major sources of training data exist for post-training modern language models: online (model-generated rollouts) data, and offline (human or other-model demonstrations) data. These two types of data are typically used by approaches like Reinforcement Learning (RL) and Supervised Fine-Tuning (SFT), respectively. In this paper, we show that these approaches are not in contradiction, but are instances of a single optimization process. We derive a Unified Policy Gradient Estimator, and present t

Source

http://arxiv.org/abs/2509.04419v2