SOURCE-LINKED INTELLIGENCE

Finetuning with Sampling: SFT Learns Better Than You Think

arXiv · AI, language, vision and robotics · article · Oct 1, 2026 · UTC

Introducing new capabilities to frontier models has long been the goal of posttraining, which predominantly employs supervised finetuning (SFT) and reinforcement learning (RL) to this end. Conventional wisdom dictates that RL enables strong generalization on new tasks without losing existing capabilities, while SFT is prone to weak generalization and catastrophic forgetting. At the same time, SFT can learn from off-policy expert data, whereas RL must rely on a model's ability to find successful trajectories with repeated sampling. In our work, we seek to leverage the strength of on-policy lear

Read original source ↗ Open in workspace

recordType

paper

region

Global