English

WST: Weak-to-Strong Knowledge Transfer via Reinforcement Learning

Machine Learning 2025-08-26 v1 Artificial Intelligence

Abstract

Effective prompt engineering remains a challenging task for many applications. We introduce Weak-to-Strong Transfer (WST), an automatic prompt engineering framework where a small "Teacher" model generates instructions that enhance the performance of a much larger "Student" model. Unlike prior work, WST requires only a weak teacher, making it efficient and broadly applicable in settings where large models are closed-source or difficult to fine-tune. Using reinforcement learning, the Teacher Model's instructions are iteratively improved based on the Student Model's outcomes, yielding substantial gains across reasoning (MATH-500, GSM8K) and alignment (HH-RLHF) benchmarks - 98% on MATH-500 and 134% on HH-RLHF - and surpassing baselines such as GPT-4o-mini and Llama-70B. These results demonstrate that small models can reliably scaffold larger ones, unlocking latent capabilities while avoiding misleading prompts that stronger teachers may introduce, establishing WST as a scalable solution for efficient and safe LLM prompt refinement.

Keywords

Cite

@article{arxiv.2508.16741,
  title  = {WST: Weak-to-Strong Knowledge Transfer via Reinforcement Learning},
  author = {Haosen Ge and Shuo Li and Lianghuan Huang},
  journal= {arXiv preprint arXiv:2508.16741},
  year   = {2025}
}
R2 v1 2026-07-01T05:02:23.549Z