Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

LegacyTranslate: LLM-based Multi-Agent Method for Legacy Code Translation

Zahra Moti, Heydar Soudani, Jonck van der Kogel

Abstract

Modernizing large legacy systems remains a major challenge in enterprise environments, particularly when migration must preserve domain-specific logic while conforming to internal architectural frameworks and shared APIs. Direct application of Large Language Models (LLMs) for code translation often produces syntactically valid outputs that fail to compile or integrate within existing production frameworks, limiting their practical adoption in real-world modernization efforts. In this paper, we propose LegacyTranslate, a multi-agent framework for API-aware code translation, developed and evaluated in the context of an ongoing modernization effort at a financial institution migrating approximately 2.5 million lines of PL/SQL to Java. The core idea is to use specialized LLM-based agents, each addressing a different aspect of the translation challenge. Specifically, LegacyTranslate consists of three agents: Initial Translation Agent produces an initial Java translation using retrieved in-context examples; API Grounding Agent aligns the code with existing APIs by retrieving relevant entries from an API knowledge base; and Refinement Agent iteratively refines the output using compiler feedback and API suggestions to improve correctness. Our experiments show that each agent contributes to better translation quality. The Initial Translation Agent alone achieves 45.6% compilable outputs and 30.9% test-pass rate. With API Grounding Agent and Refinement Agent, compilation improves by an additional 8% and test-pass accuracy increases by 3%.

LegacyTranslate: LLM-based Multi-Agent Method for Legacy Code Translation

Abstract

Modernizing large legacy systems remains a major challenge in enterprise environments, particularly when migration must preserve domain-specific logic while conforming to internal architectural frameworks and shared APIs. Direct application of Large Language Models (LLMs) for code translation often produces syntactically valid outputs that fail to compile or integrate within existing production frameworks, limiting their practical adoption in real-world modernization efforts. In this paper, we propose LegacyTranslate, a multi-agent framework for API-aware code translation, developed and evaluated in the context of an ongoing modernization effort at a financial institution migrating approximately 2.5 million lines of PL/SQL to Java. The core idea is to use specialized LLM-based agents, each addressing a different aspect of the translation challenge. Specifically, LegacyTranslate consists of three agents: Initial Translation Agent produces an initial Java translation using retrieved in-context examples; API Grounding Agent aligns the code with existing APIs by retrieving relevant entries from an API knowledge base; and Refinement Agent iteratively refines the output using compiler feedback and API suggestions to improve correctness. Our experiments show that each agent contributes to better translation quality. The Initial Translation Agent alone achieves 45.6% compilable outputs and 30.9% test-pass rate. With API Grounding Agent and Refinement Agent, compilation improves by an additional 8% and test-pass accuracy increases by 3%.
Paper Structure (9 sections, 4 equations, 2 figures, 2 tables)

This paper contains 9 sections, 4 equations, 2 figures, 2 tables.

Table of Contents

  1. Introduction
  2. Related work
  3. Methodology
  4. Initial Translation Agent
  5. API Grounding Agent
  6. Refinement Agent
  7. Experimental Setup
  8. Results
  9. Conclusions and Future Work

Figures (2)

  • Figure 1: Overview of LegacyTranslate for PL/SQL$\rightarrow$Java translation. (a) Initial Translation Agent takes a test sample and reference set and generates an initial code using in-context examples retrieval. (b) The code is evaluated using unit tests. (c) API Grounding Agent shortlists relevant APIs from a knowledge base. (d) Refinement Agent iteratively corrects the code using API details and execution feedback. Solid lines = standard flow; dotted = executed only on failure.
  • Figure 2: Performance of different retrieval models on retrieving examples in Initial Translation Agent.