Towards Mitigating API Hallucination in Code Generated by LLMs with Hierarchical Dependency Aware

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chen, Yujia, Chen, Mingyu, Gao, Cuiyun, Jiang, Zhihan, Li, Zhongqi, Ma, Yuchi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913848557043712
author Chen, Yujia
Chen, Mingyu
Gao, Cuiyun
Jiang, Zhihan
Li, Zhongqi
Ma, Yuchi
author_facet Chen, Yujia
Chen, Mingyu
Gao, Cuiyun
Jiang, Zhihan
Li, Zhongqi
Ma, Yuchi
contents Application Programming Interfaces (APIs) are crucial in modern software development. Large Language Models (LLMs) assist in automated code generation but often struggle with API hallucination, including invoking non-existent APIs and misusing existing ones in practical development scenarios. Existing studies resort to Retrieval-Augmented Generation (RAG) methods for mitigating the hallucination issue, but tend to fail since they generally ignore the structural dependencies in practical projects and do not indeed validate whether the generated APIs are available or not. To address these limitations, we propose MARIN, a framework for mitigating API hallucination in code generated by LLMs with hierarchical dependency aware. MARIN consists of two phases: Hierarchical Dependency Mining, which analyzes local and global dependencies of the current function, aiming to supplement comprehensive project context in LLMs input, and Dependency Constrained Decoding, which utilizes mined dependencies to adaptively constrain the generation process, aiming to ensure the generated APIs align with the projects specifications. To facilitate the evaluation of the degree of API hallucination, we introduce a new benchmark APIHulBench and two new metrics including Micro Hallucination Number (MiHN) and Macro Hallucination Rate (MaHR). Experiments on six state-of-the-art LLMs demonstrate that MARIN effectively reduces API hallucinations, achieving an average decrease of 67.52% in MiHN and 73.56% in MaHR compared to the RAG approach. Applied to Huaweis internal projects and two proprietary LLMs, MARIN achieves average decreases of 57.33% in MiHN and 59.41% in MaHR.
format Preprint
id arxiv_https___arxiv_org_abs_2505_05057
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Towards Mitigating API Hallucination in Code Generated by LLMs with Hierarchical Dependency Aware
Chen, Yujia
Chen, Mingyu
Gao, Cuiyun
Jiang, Zhihan
Li, Zhongqi
Ma, Yuchi
Software Engineering
Application Programming Interfaces (APIs) are crucial in modern software development. Large Language Models (LLMs) assist in automated code generation but often struggle with API hallucination, including invoking non-existent APIs and misusing existing ones in practical development scenarios. Existing studies resort to Retrieval-Augmented Generation (RAG) methods for mitigating the hallucination issue, but tend to fail since they generally ignore the structural dependencies in practical projects and do not indeed validate whether the generated APIs are available or not. To address these limitations, we propose MARIN, a framework for mitigating API hallucination in code generated by LLMs with hierarchical dependency aware. MARIN consists of two phases: Hierarchical Dependency Mining, which analyzes local and global dependencies of the current function, aiming to supplement comprehensive project context in LLMs input, and Dependency Constrained Decoding, which utilizes mined dependencies to adaptively constrain the generation process, aiming to ensure the generated APIs align with the projects specifications. To facilitate the evaluation of the degree of API hallucination, we introduce a new benchmark APIHulBench and two new metrics including Micro Hallucination Number (MiHN) and Macro Hallucination Rate (MaHR). Experiments on six state-of-the-art LLMs demonstrate that MARIN effectively reduces API hallucinations, achieving an average decrease of 67.52% in MiHN and 73.56% in MaHR compared to the RAG approach. Applied to Huaweis internal projects and two proprietary LLMs, MARIN achieves average decreases of 57.33% in MiHN and 59.41% in MaHR.
title Towards Mitigating API Hallucination in Code Generated by LLMs with Hierarchical Dependency Aware
topic Software Engineering
url https://arxiv.org/abs/2505.05057