# Enhancing COBOL Code Explanations: A Multi-Agents Approach Using Large Language Models

arXiv: https://arxiv.org/abs/2507.02182 · HTML: https://mlsanity.com/paper/2507.02182/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 17 of its 55 references that are in the map.
- First posted on arXiv 2 Jul 2025.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Builds on (most cited first, 17 of 17)

- [Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks](https://mlsanity.com/paper/1908.10084/) (2019)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context](https://mlsanity.com/paper/2403.05530/) (2024)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [JudgeBench: A Benchmark for Evaluating LLM-based Judges](https://mlsanity.com/paper/2410.12784/) (2024)
- [Recursively Summarizing Books with Human Feedback](https://mlsanity.com/paper/2109.10862/) (2021)
- [BooookScore: A systematic exploration of book-length summarization in the era of LLMs](https://mlsanity.com/paper/2310.00785/) (2023)
- [Granite Code Models: A Family of Open Foundation Models for Code Intelligence](https://mlsanity.com/paper/2405.04324/) (2024)
- [Automatic Semantic Augmentation of Language Model Prompts (for Code Summarization)](https://mlsanity.com/paper/2304.06815/) (2023)
- [Distilled GPT for Source Code Summarization](https://mlsanity.com/paper/2308.14731/) (2023)
- [GPTutor: a ChatGPT-powered programming tool for code explanation](https://mlsanity.com/paper/2305.01863/) (2023)
- [Legal Case Document Similarity: You Need Both Network and Text](https://mlsanity.com/paper/2209.12474/) (2022)
- [XMainframe: A Large Language Model for Mainframe Modernization](https://mlsanity.com/paper/2408.04660/) (2024)
- [Including Everyone, Everywhere: Understanding Opportunities and Challenges of Geographic Gender-Inclusion in OSS](https://mlsanity.com/paper/2010.00822/) (2020)
- [X-COBOL: A Dataset of COBOL Repositories](https://mlsanity.com/paper/2306.04892/) (2023)
- [Contemporary COBOL: Developers' Perspectives on Defects and Defect Location](https://mlsanity.com/paper/2105.01830/) (2021)
- [An Empirical Study On Correlation between Readme Content and Project Popularity](https://mlsanity.com/paper/2206.10772/) (2022)

## Built on by (most cited first, 2 of 2)

- [Evaluating perturbation robustness of generative systems that use COBOL code inputs](https://mlsanity.com/paper/2511.18488/) (2025)
- [Spec2COBOLRot: An Agentic-AI Degradation Loop for Realistic COBOL Corpus Generation](https://mlsanity.com/paper/2609.26835/) (2026)

## Abstract (arXiv)

Common Business Oriented Language (COBOL) is a programming language used to develop business applications that are widely adopted by financial, business, and government agencies. Due to its age, complexity, and declining number of COBOL developers, maintaining COBOL codebases is becoming increasingly challenging. In particular, the lack of documentation makes it difficult for new developers to effectively understand and maintain COBOL systems. Existing research utilizes large language models (LLMs) to explain the functionality of code snippets. However, COBOL presents unique challenges due to its architectural and syntactical differences, which often cause its code to exceed the token window size of LLMs. In this work, we propose a multi-agent approach that leverages two LLM-based agents working collaboratively to generate explanations for functions, files, and the overall project. These agents incorporate together by utilizing contextual information from the codebase into the code explanation prompts. We evaluate the effectiveness of our approach using 14 open-source, real-world COBOL projects. Our results indicate that our approach performs significantly better than the baseline in function code explanation, with improvements of 12.67%, 18.59%, and 0.62% in terms of METEOR, chrF, and SentenceBERT scores, respectively. At the file level, our approach effectively explains both short and long COBOL files that exceed the token window size of LLMs and surpass the baseline by 4.21%, 10.72%, and 14.68% in explaining the purpose, functionality, and clarity of the generated explanation. At the project level, our approach generates explanations that convey the functionality and purpose of 82% of the selected projects.