How Robust are LLM-Generated Library Imports? An Empirical Study using Stack Overflow

How Robust are LLM-Generated Library Imports? Abstract --Software libraries are central to the functionality, security, and maintainability of modern code. As developers increasingly turn to Large Language Models (LLMs) to assist with programming tasks, understanding how these models recommend libraries is essential. In this paper, we conduct an empirical study of six state-of-the-art LLMs, both proprietary and open-source, by prompting them to solve real-world Python problems sourced from Stack Overflow. We analyze the types of libraries they import, the characteristics of those libraries, and the extent to which the recommendations are usable out of the box. Our results show that LLMs predominantly favour third-party libraries over standard ones, and often recommend mature, popular, and permissively licensed dependencies. However, we also identify gaps in usability: 4.6% of the libraries could not be resolved automatically due to structural mismatches between import names and installable packages, and only two models (out of six) provided installation guidance. While the generated code is technically valid, the lack of contextual support places the burden of manually resolving dependencies on the user . Our findings offer actionable insights for both developers and researchers, and highlight opportunities to improve the reliability and usability of LLM-generated code in the context of software dependencies. ODERN software development heavily relies on open source libraries that provide reusable functionalities through well-defined modules, significantly reducing development time and effort [1]-[3]. While libraries can help speed up development tasks, they also introduce dependencies -- interconnections between code components -- that can lead to increased complexity and dependency management challenges [4]-[6]. One critical aspect of dependency management is library selection [7]. Previous studies have explored how developers select libraries, and highlighted primarily ad-hoc processes based on past experiences, expert advice, and online resources [8], [9]. Decisions around dependency adoption are influenced by factors such as functionality, community support, and maintenance compatibility [7], [10], [11]. In parallel, the growing adoption of LLMs as programming assistants introduces new possibilities for addressing these challenges. LLMs are increasingly used to assist with code generation, and studies show their potential to enhance productivity through capabilities like code completion and search [24]-[26]. However, their impact on software dependencies (i.e., their ability to generate reliable library imports) remains unexplored.