# Style2Code: A Style-Controllable Code Generation Framework with Dual-Modal Contrastive Representation Learning

arXiv: https://arxiv.org/abs/2505.19442 · HTML: https://mlsanity.com/paper/2505.19442/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 9 of its 16 references that are in the map.
- First posted on arXiv 26 May 2025.

## Builds on (most cited first, 9 of 9)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Abstract Syntax Networks for Code Generation and Semantic Parsing](https://mlsanity.com/paper/1704.07535/) (2017)
- [Automatic Semantic Augmentation of Language Model Prompts (for Code Summarization)](https://mlsanity.com/paper/2304.06815/) (2023)
- [Out of the BLEU: how should we assess quality of the Code Generation models?](https://mlsanity.com/paper/2208.03133/) (2022)
- [Self-Supervised Contrastive Learning for Code Retrieval and Summarization via Semantic-Preserving Transformations](https://mlsanity.com/paper/2009.02731/) (2020)
- [A Review of Text Style Transfer using Deep Learning](https://mlsanity.com/paper/2109.15144/) (2021)
- [Code-Style In-Context Learning for Knowledge-Based Question Answering](https://mlsanity.com/paper/2309.04695/) (2023)
- [MPCODER: Multi-user Personalized Code Generator with Explicit and Implicit Style Representation Learning](https://mlsanity.com/paper/2406.17255/) (2024)
- [I Can Find You in Seconds! Leveraging Large Language Models for Code Authorship Attribution](https://mlsanity.com/paper/2501.08165/) (2025)

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

- [Themis: Training Robust Multilingual Code Reward Models for Flexible Multi-Criteria Scoring](https://mlsanity.com/paper/2605.00754/) (2026)

## Abstract (arXiv)

Controllable code generation, the ability to synthesize code that follows a specified style while maintaining functionality, remains a challenging task. We propose a two-stage training framework combining contrastive learning and conditional decoding to enable flexible style control. The first stage aligns code style representations with semantic and structural features. In the second stage, we fine-tune a language model (e.g., Flan-T5) conditioned on the learned style vector to guide generation. Our method supports style interpolation and user personalization via lightweight mixing. Compared to prior work, our unified framework offers improved stylistic control without sacrificing code correctness. This is among the first approaches to combine contrastive alignment with conditional decoding for style-guided code generation.