paper-with-me

홈 › Papers

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

2025-08-22 · Haosen Ge, Shuo Li, Lianghuan Huang arxiv

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.

📄 PDF Abstract BibTeX arXiv:2508.16741

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement LearningPrompt Engineering

Similar Papers 제목 키워드 기반

A transfer learning framework for weak-to-strong generalization

2024-05-25 · Seamus Somerstep, Felipe Maia Polo, Moulinath Banerjee, Ya'acov Ritov 외

Modern large language model (LLM) alignment techniques rely on human feedback, but it is unclear whether these techniques fundamentally limit the capabilities of aligned LLMs. In particular, it is unknown if it is possib…

Language ModelingLanguage ModellingLarge Language ModelTransfer Learning

Detector Discovery in the Wild: Joint Multiple Instance and Representation Learning

2014-12-02 · CVPR 2015 6 · Judy Hoffman, Deepak Pathak, Trevor Darrell, Kate Saenko

We develop methods for detector learning which exploit joint training over both weak and strong labels and which transfer learned perceptual representations from strongly-labeled auxiliary tasks. Previous methods for wea…

Multiple Instance LearningRepresentation LearningTransfer Learning

Transferable Model-agnostic Vision-Language Model Adaptation for Efficient Weak-to-Strong Generalization

2025-08-12 · Jihwan Park, Taehoon Song, Sanghyeok Lee, Miso Choi 외 arxiv

Vision-Language Models (VLMs) have been widely used in various visual recognition tasks due to their remarkable generalization capabilities. As these models grow in size and complexity, fine-tuning becomes costly, emphas…

Can a powerful neural network be a teacher for a weaker neural network?

2020-05-01 · Nicola Landro, Ignazio Gallo, Riccardo La Grassa

The transfer learning technique is widely used to learning in one context and applying it to another, i.e. the capacity to apply acquired knowledge and skills to new situations. But is it possible to transfer the learnin…

Transfer Learning

Combining inherent knowledge of vision-language models with unsupervised domain adaptation through strong-weak guidance

2023-12-07 · Thomas Westfechtel, Dexuan Zhang, Tatsuya Harada

Unsupervised domain adaptation (UDA) tries to overcome the tedious work of labeling data by leveraging a labeled source dataset and transferring its knowledge to a similar but different target dataset. Meanwhile, current…

Domain AdaptationKnowledge DistillationLanguage ModellingSelf-Knowledge Distillation+1