Scaling Near-Optimal SFT-RL Annotation Budget Allocation from Small to Large LLMs

Paper recorded by Signals 4 on 2026-09-01 in cs.AI. Abstract reproduced from arXiv; link to the original below.

Published 2026-09-01 on arXiv · recorded by Signals 4 on 2026-09-02

Category: cs.AI · 人工智能 · first seen 2026-09-02

Abstract

How to divide a fixed annotation budget between supervised fine-tuning (SFT) and reinforcement learning (RL) during LLM post-training remains an open problem. Existing work characterizes only broad trends (e.g., SFT dominates in low-data regimes), lacks a principled allocation framework, and does not examine whether the optimal ratio transfers across model sizes. We frame this problem in terms of

Read on arXiv →