Show and Tell: Prompt Strategies for Style Control in Multi-Turn LLM Code Generation

TL;DR

This paper investigates how different prompt strategies influence style control in multi-turn LLM code generation, revealing that combined prompts offer the most stable stylistic control during iterative code refinement.

Contribution

It systematically compares instruction, example, and combined prompts in a two-turn code generation task, highlighting their distinct effects on style control and expansion discipline.

Findings

Combined prompts yield strongest initial style control and expansion discipline.

Instruction prompts have large initial effects but moderate expansion control.

Example prompts show modest initial effects with no expansion discipline.

Abstract

Language models generate functionally correct code that tends toward excessive verbosity, with elaborate documentation and defensive patterns that diverge from human baselines. Two prompting mechanisms have emerged for stylistic control: instruction based prompts that articulate abstract directives, and example based prompts that provide concrete code demonstrations. The core problem is whether stylistic constraints persist when models enhance initial implementations with additional features while maintaining high functional accuracy. Here we show that instruction-based, example-based, and combined prompts produce distinct patterns of initial control and expansion discipline over one enhancement turn. We manipulated system prompts across four conditions in a paired two-turn protocol where models first generated solutions to an intermediate Python task, then revised their code under general improvement directives, holding the user task fixed (N = 160 paired programs). Combined prompts produced the strongest initial compression and greatest expansion discipline. Instructions showed large initial effects and moderate expansion discipline. Examples showed modest initial effects with no expansion discipline. These results show that initial prompt effectiveness and expansion discipline are separate aspects of prompt design, and that combined approaches provide the most stable stylistic control in this two-turn workflow.