PromptTailor: Multi-turn Intent-Aligned Prompt Synthesis for Lightweight LLMs

TL;DR

PromptTailor is a lightweight system that generates intent-aligned prompts for small language models, improving response quality and user alignment with fewer model calls through fine-tuned prompt synthesis.

Contribution

It introduces a lightweight, fine-tuned prompt synthesis method for LLMs that enhances output quality and alignment with user intent, suitable for edge deployment.

Findings

Outperforms chain-of-thought prompting in preference tests

Matches or exceeds state-of-the-art prompt optimization methods

Requires fewer model calls (e.g., 3 vs. 9) for effective prompt generation

Abstract

Lightweight language models remain attractive for on-device and privacy-sensitive applications, but their responses are highly sensitive to prompt quality. For open-ended generation, non-expert users often lack the knowledge or time to consistently craft high-quality prompts, leading them to rely on prompt optimization tools. However, a key challenge is ensuring the optimized prompts genuinely align with users' original intents and preferences. We introduce PromptTailor, a system for controllable prompt generation for open-ended text that improves model output quality by intent-aligned prompt synthesis. PromptTailor expands minimal user instructions into rich, domain-aware prompts while preserving the user's stated preferences. The system is a quantized Llama3-8B model fine-tuned with a lightweight LoRA adapter on 12,300 prompt-refinement dialogues spanning 41 everyday domains, distilled from three stronger LLMs. The adapter attaches to any Llama3-8B base, enabling edge deployment. In human and LLM-judge evaluations across multiple target models and optimization baselines, PromptTailor yields higher preference rates than chain-of-thought prompting and matches or surpasses state-of-the-art prompt optimization methods while requiring fewer model calls (e.g., 3 vs. 9). These results show that a compact student, guided by powerful teachers, can learn effective prompt-generation strategies that enhance response quality while maintaining alignment with user intent.