# Collaboration is all you need: LLM Assisted Safe Code Translation

arXiv: https://arxiv.org/abs/2503.11237 · HTML: https://mlsanity.com/paper/2503.11237/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 17 references that are in the map.
- First posted on arXiv 14 Mar 2025.

## Builds on (most cited first, 7 of 7)

- [Unsupervised Translation of Programming Languages](https://mlsanity.com/paper/2006.03511/) (2020)
- [AVATAR: A Parallel Corpus for Java-Python Program Translation](https://mlsanity.com/paper/2108.11590/) (2021)
- [XLCoST: A Benchmark Dataset for Cross-lingual Code Intelligence](https://mlsanity.com/paper/2206.08474/) (2022)
- [Understanding the Performance and Estimating the Cost of LLM Fine-Tuning](https://mlsanity.com/paper/2408.04693/) (2024)
- [SteloCoder: a Decoder-Only LLM for Multi-Language to Python Code Translation](https://mlsanity.com/paper/2310.15539/) (2023)
- [CARGO: AI-Guided Dependency Analysis for Migrating Monolithic Applications to Microservices Architecture](https://mlsanity.com/paper/2207.11784/) (2022)
- [Teaching Machines to Code: Smart Contract Translation with LLMs](https://mlsanity.com/paper/2403.09740/) (2024)

## Built on by (most cited first, 1 of 1)

- [Revisiting the Role of Natural Language Code Comments in Code Translation](https://mlsanity.com/paper/2601.16661/) (2026)

## Abstract (arXiv)

This paper introduces UniTranslator, a visionary framework that re-imagines code translation as a collaborative endeavor among multiple, compact LLMs. By orchestrating the interaction of specialized agents, each focused on different aspects of the translation process and grounded in a deep understanding of programming concepts, UniTranslator achieves a level of accuracy and efficiency that rivals larger, monolithic models. Our preliminary evaluation demonstrates the potential of UniTranslator to overcome the limitations of existing approaches and unlock the power of smaller LLMs for complex code translation tasks. We explore the effectiveness of this dynamic multi-agent paradigm in handling diverse language pairs, including low-resource languages, and in mitigating common issues such as code artifacts and hallucinations through the use of Natural Language Inference (NLI) grounding and iterative feedback mechanisms