# UnitCoder: Scalable Iterative Code Synthesis with Unit Test Guidance

arXiv: https://arxiv.org/abs/2502.11460 · HTML: https://mlsanity.com/paper/2502.11460/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 17 Feb 2025.

## Built on by (most cited first, 7 of 7)

- [Benchmark Everything Everywhere All at Once](https://mlsanity.com/paper/2606.06462/) (2026)
- [CodeEvo: Interaction-Driven Synthesis of Code-centric Data through Hybrid and Iterative Feedback](https://mlsanity.com/paper/2507.22080/) (2025)
- [CodeScaler: Scaling Code LLM Training and Test-Time Inference via Reward Models](https://mlsanity.com/paper/2602.17684/) (2026)
- [Code-driven Number Sequence Calculation: Enhancing the inductive Reasoning Abilities of Large Language Models](https://mlsanity.com/paper/2510.14620/) (2025)
- [P4OMP: Retrieval-Augmented Prompting for OpenMP Parallelism in Serial Code](https://mlsanity.com/paper/2506.22703/) (2025)
- [Knowledge-Graph-Driven Data Synthesis for Low-Resource Software Development: A HarmonyOS Case Study](https://mlsanity.com/paper/2512.00380/) (2025)
- [RTLCurator: Label-Efficient Data Curation for RTL Generation](https://mlsanity.com/paper/2607.29283/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) have demonstrated remarkable capabilities in various tasks, yet code generation remains a major challenge. Current approaches for obtaining high-quality code data primarily focus on (i) collecting large-scale pre-training data and (ii) synthesizing instruction data through prompt engineering with powerful models. While pre-training data faces quality consistency issues, instruction-based synthesis suffers from limited instruction diversity and inherent biases of LLMs. To address this gap, we introduce UnitCoder, a systematic pipeline leveraging model-generated unit tests to both guide and validate the code generation process. Combined with large-scale package-based retrieval from pre-training corpus, we generate a dataset of 500K+ verifiable programs containing diverse API calls. Evaluations on multiple Python benchmarks (BigCodeBench, HumanEval, MBPP) demonstrate that models fine-tuned on our synthetic data exhibit consistent performance improvements. Notably, Llama3.1-8B and InternLM2.5-7B improve from 31% and 28% to 40% and 39% success rates on BigCodeBench, respectively. Our work presents a scalable approach that leverages model-generated unit tests to guide the synthesis of high-quality code data from pre-training corpora, demonstrating the potential for producing diverse and high-quality post-training data at scale. All code and data will be released (https://github.com).