Learning to Rewrite Prompts for Bootstrapping LLMs on Downstream Tasks

2025/10/08 by Qinhao Zhou, Zhou, Qinhao, Xiang Xiang +4
Computer Science · #Artificial Intelligence (cs.AI) #Audio and Speech Processing (eess.AS) #Computation and Language (cs.CL) #FOS: Computer and information sciences #FOS: Electrical engineering #Machine Learning (cs.LG) #Natural Language Processing Techniques #Semantic Web and Ontologies #Topic Modeling #electronic engineering #information engineering

paper · pdf · doi:10.48550/arxiv.2510.06695

openalex publication_date 2025/10/08 · openalex created_date 2025/10/13 · openalex updated_date 2026/07/28

Abstract

In recent years, the growing interest in Large Language Models (LLMs) has significantly advanced prompt engineering, transitioning from manual design to model-based optimization. Prompts for LLMs generally comprise two components: the instruction, which defines the task or objective, and the input, which is tailored to the instruction type. In natural language generation (NLG) tasks such as machine translation, the input component is particularly critical, while the instruction component tends to be concise. Existing prompt engineering methods primarily focus on optimizing the instruction component for general tasks, often requiring large-parameter LLMs as auxiliary tools. However, these approaches exhibit limited applicability for tasks like machine translation, where the input component plays a more pivotal role. To address this limitation, this paper introduces a novel prompt optimization method specifically designed for machine translation tasks. The proposed approach employs a small-parameter model trained using a back-translation-based strategy, significantly reducing training overhead for single-task optimization while delivering highly effective performance. With certain adaptations, this method can also be extended to other downstream tasks.

Citations

• Learning to Rewrite: Generalized LLM-Generated Text Detection
• Prompt Optimization with Human Feedback
• PRewrite: Prompt Rewriting with Reinforcement Learning
• Qwen Technical Report
• Automatic Prompt Optimization with "Gradient Descent" and Beam Search
• LLaMA: Open and Efficient Foundation Language Models
• Augmented Language Models: a Survey
• Demystifying Prompts in Language Models via Perplexity Estimation
• Large Language Models Are Human-Level Prompt Engineers
• RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learning
• PaLM: Scaling Language Modeling with Pathways
• BNAI, NO-TOKEN, and MIND-UNITY: Pillars of a Systemic Revolution in Artificial Intelligence
• Learning To Retrieve Prompts for In-Context Learning
• P-Tuning v2: Prompt Tuning Can Be Comparable to Fine-tuning Universally Across Scales and Tasks
• Finetuned Language Models Are Zero-Shot Learners
• The Power of Scale for Parameter-Efficient Prompt Tuning
• GPT Understands, Too
• GLM: General Language Model Pretraining with Autoregressive Blank Infilling
• What Makes Good In-Context Examples for GPT-3?
• mT5: A massively multilingual pre-trained text-to-text transformer
• Language Models are Few-Shot Learners
• Multilingual Denoising Pre-training for Neural Machine Translation
• Don't Give Me the Details, Just the Summary! Topic-Aware Convolutional\n Neural Networks for Extreme Summarization
• GLUE: A Multi-Task Benchmark and Analysis Platform for Natural Language Understanding
• Prefix-Tuning: Optimizing Continuous Prompts for Generation
• Prefix-Tuning: Optimizing Continuous Prompts for Generation

Related

• MAPO: Boosting Large Language Model Performance with Model-Adaptive Prompt Optimization
• SEE: Strategic Exploration and Exploitation for Cohesive In-Context Prompt Optimization
• Large Language Models Are Human-Level Prompt Engineers
• A Survey of Prompt Engineering Methods in Large Language Models for Different NLP Tasks
• A Systematic Survey of Automatic Prompt Optimization Techniques