---
title: "ReSyn: A Generalized Recursive Regular Expression Synthesis Framework"
authors:
  - "Seongmin Kim"
  - "Hyunjoon Cheon"
  - "Su-Hyeon Kim"
  - "Yo-Sub Han"
  - "Sang-Ki Ko"
doi: "10.24963/ijcai.2026/435"
url: "https://ijcai.org/proceedings/2026/435"
pdf: "https://ijcai.org/proceedings/2026/0435.pdf"
venue: "Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence"
year: 2026
pages: "3908-3916"
track: "Main Track"
keywords:
  - "Knowledge Representation and Reasoning: Knowledge representation languages"
  - "Knowledge Representation and Reasoning: Computational complexity of reasoning"
  - "Knowledge Representation and Reasoning: Learning and reasoning"
  - "Knowledge Representation and Reasoning: Automated reasoning and theorem proving"
---

# ReSyn: A Generalized Recursive Regular Expression Synthesis Framework

Seongmin Kim, Hyunjoon Cheon, Su-Hyeon Kim, Yo-Sub Han, Sang-Ki Ko

*Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence*, pages 3908-3916. DOI: [10.24963/ijcai.2026/435](https://doi.org/10.24963/ijcai.2026/435)

## Abstract

Existing Programming-By-Example (PBE) systems often rely on simplified benchmarks that fail to capture the high structural complexity of real-world regexes, such as deeper nesting and frequent use of union operations.
To overcome the resulting performance drop, we propose ReSyn, a synthesizer-agnostic divide-and-conquer framework that decomposes complex synthesis problem into manageable sub-problems.
We also introduce Set2Regex, a parameter-efficient synthesizer capturing the permutation invariance of examples.
Experimental results demonstrate that ReSyn significantly boosts accuracy across various synthesizers, and its combination with Set2Regex establishes a new state-of-the-art on challenging real-world benchmark.
The complete source code, datasets, and pre-trained model checkpoints are publicly available at https://github.com/mrseongminkim/ReSyn.