A Causal Perspective on Measuring, Explaining and Mitigating Smells in LLM-Generated Code

TL;DR

This paper introduces a causal analysis framework using the Propensity Smelly Score (PSC) to understand, measure, and reduce code smells in LLM-generated code, enhancing code quality and developer interpretability.

Contribution

It proposes a novel causal approach with PSC to analyze factors influencing code smells and offers mitigation strategies to improve LLM-generated code quality.

Findings

Prompt design significantly impacts smell propensity.

Model architecture influences code quality.

PSC aids developers in interpreting model behavior.

Abstract

Recent advances in large language models (LLMs) have accelerated their adoption in software engineering contexts. However, concerns persist about the structural quality of the code they produce. In particular, LLMs often replicate poor coding practices, introducing code smells (i.e., patterns that hinder readability, maintainability, or design integrity). Although prior research has examined the detection or repair of smells, we still lack a clear understanding of how and when these issues emerge in generated code. This paper addresses this gap by systematically measuring, explaining and mitigating smell propensity in LLM-generated code. We build on the Propensity Smelly Score (PSC), a probabilistic metric that estimates the likelihood of generating particular smell types, and establish its robustness as a signal of structural quality. Using PSC as an instrument for causal analysis, we identify how generation strategy, model size, model architecture and prompt formulation shape the structural properties of generated code. Our findings show that prompt design and architectural choices play a decisive role in smell propensity and motivate practical mitigation strategies that reduce its occurrence. A user study further demonstrates that PSC helps developers interpret model behavior and assess code quality, providing evidence that smell propensity signals can support human judgement. Taken together, our work lays the groundwork for integrating quality-aware assessments into the evaluation and deployment of LLMs for code.