Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.14281 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914044190916608 |
|---|---|
| author | Yao, Xifeng Lang, Dongyu Zhang, Wu Guo, Xintong Xie, Huarui Ni, Yinhao Liu, Ping Shen, Guang Bai, Yi Tu, Dandan Zhang, Changzheng |
| author_facet | Yao, Xifeng Lang, Dongyu Zhang, Wu Guo, Xintong Xie, Huarui Ni, Yinhao Liu, Ping Shen, Guang Bai, Yi Tu, Dandan Zhang, Changzheng |
| contents | Significant advancements have been made in the capabilities of code large language models, leading to their rapid adoption and application across a wide range of domains. However, their further advancements are often constrained by the scarcity of real-world coding problems. To bridge this gap, we propose a novel framework for synthesizing code problems that emulate authentic real-world scenarios. This framework systematically integrates domain knowledge, domain skills, and coding skills, all of which are meticulously extracted from real-world programming-related datasets, including Stack Overflow and Kaggle. The extracted elements serve as the foundational building blocks for constructing code problems. To align the generated problems with practical applications, application scenarios are also mined from the aforementioned datasets. These scenarios are then utilized to construct a scenario-centric graph that interconnects domain knowledge, domain skills, and coding skills. Based on this structured representation, a sampling strategy on the graph is designed, which effectively controls the generation of a code problem with complexity and diversity, reflects real-world challenges. Experimental results demonstrate that the proposed method consistently achieves superior performance over state-of-the-art open-source large language models of varying sizes and functionalities, including both coders and general-purpose models, across a diverse set of real-world benchmarks. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14281 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SCoGen: Scenario-Centric Graph-Based Synthesis of Real-World Code Problems Yao, Xifeng Lang, Dongyu Zhang, Wu Guo, Xintong Xie, Huarui Ni, Yinhao Liu, Ping Shen, Guang Bai, Yi Tu, Dandan Zhang, Changzheng Software Engineering Artificial Intelligence Significant advancements have been made in the capabilities of code large language models, leading to their rapid adoption and application across a wide range of domains. However, their further advancements are often constrained by the scarcity of real-world coding problems. To bridge this gap, we propose a novel framework for synthesizing code problems that emulate authentic real-world scenarios. This framework systematically integrates domain knowledge, domain skills, and coding skills, all of which are meticulously extracted from real-world programming-related datasets, including Stack Overflow and Kaggle. The extracted elements serve as the foundational building blocks for constructing code problems. To align the generated problems with practical applications, application scenarios are also mined from the aforementioned datasets. These scenarios are then utilized to construct a scenario-centric graph that interconnects domain knowledge, domain skills, and coding skills. Based on this structured representation, a sampling strategy on the graph is designed, which effectively controls the generation of a code problem with complexity and diversity, reflects real-world challenges. Experimental results demonstrate that the proposed method consistently achieves superior performance over state-of-the-art open-source large language models of varying sizes and functionalities, including both coders and general-purpose models, across a diverse set of real-world benchmarks. |
| title | SCoGen: Scenario-Centric Graph-Based Synthesis of Real-World Code Problems |
| topic | Software Engineering Artificial Intelligence |
| url | https://arxiv.org/abs/2509.14281 |