Understanding before Naming! Enhancing LLM-based Method Name Prediction with Code Summarization

2026-07-14

AI Summary

This work addresses the limitations of existing method name prediction (MNP) approaches, which often rely on direct code-to-name mappings and employ evaluation metrics that poorly align with human judgments of semantic quality. To overcome these issues, we propose SMNP, a novel framework that emulates the human cognitive process of first comprehending a method’s functionality and then generating an appropriate name. SMNP enhances semantic quality through MNP-oriented code summarization and a chain-of-thought–based refinement strategy. Furthermore, we introduce large language models (LLMs) as evaluators that better approximate human assessment of naming quality. Extensive experiments across five LLMs and two benchmark datasets demonstrate that SMNP consistently outperforms state-of-the-art methods, exhibiting both strong effectiveness and robustness.

Abstract

Method names are critical to software quality, affecting code comprehensibility, maintainability, and developer collaboration. However, manually designing meaningful method names is challenging. Method Name Prediction (MNP), which automatically generates method names from code snippets, has recently attracted attention. Although large language models (LLMs) show promising performance for MNP, two challenges remain. First, existing evaluations mainly rely on token similarity metrics, which often fail to reflect human judgments of semantic quality. Second, current LLM-based MNP methods usually generate names through direct code-to-name mapping, which differs from the human process of understanding functionality before naming. To address these challenges, we conduct empirical studies on LLM-based evaluation and MNP strategies. We compare 6 metric-based evaluators, 5 LLM-based evaluators, and 6 human evaluators. Results show that LLM-based evaluators, especially DeepSeek-based evaluators, are more consistent with human judgments than traditional metrics. We further compare direct generation and summarization-and-refinement strategies. Results indicate that summarization and refinement generally improve the semantic quality of generated names. Case studies reveal three limitations: inaccurate summaries, semantic misalignment, and close semantic scores. Based on these findings, we propose SMNP, an MNP approach combining MNP-oriented summarization and chain-of-thought enhanced refinement. Experiments on 5 LLMs and 2 datasets demonstrate the effectiveness and robustness of SMNP.

Problem

• Method Name Prediction
• Code Summarization
• Large Language Models
• Semantic Quality
• Evaluation Metrics

Innovation

• Method Name Prediction
• Code Summarization
• Large Language Models
• Chain-of-Thought
• Semantic Evaluation