Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2508.06017 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916887383769088 |
|---|---|
| author | Xu, Xiangzhe Feng, Shiwei Su, Zian Wang, Chengpeng Zhang, Xiangyu |
| author_facet | Xu, Xiangzhe Feng, Shiwei Su, Zian Wang, Chengpeng Zhang, Xiangyu |
| contents | Intelligent coding systems are transforming software development by enabling users to specify code behavior in natural language. However, the opaque decision-making of AI-driven coders raises trust and usability concerns, particularly for non-expert users who cannot inspect low-level implementations. We argue that these systems should not only generate code but also produce clear, consistent justifications that bridge model reasoning and user understanding. To this end, we identify two critical justification properties-cognitive alignment and semantic faithfulness-and highlight the limitations of existing methods, including formal verification, static analysis, and post-hoc explainability. We advocate exploring neuro-symbolic approaches for justification generation, where symbolic constraints guide model behavior during training and program semantics are enriched through neural representations, enabling automated consistency checks at inference time. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_06017 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Position: Intelligent Coding Systems Should Write Programs with Justifications Xu, Xiangzhe Feng, Shiwei Su, Zian Wang, Chengpeng Zhang, Xiangyu Software Engineering Computation and Language Machine Learning Intelligent coding systems are transforming software development by enabling users to specify code behavior in natural language. However, the opaque decision-making of AI-driven coders raises trust and usability concerns, particularly for non-expert users who cannot inspect low-level implementations. We argue that these systems should not only generate code but also produce clear, consistent justifications that bridge model reasoning and user understanding. To this end, we identify two critical justification properties-cognitive alignment and semantic faithfulness-and highlight the limitations of existing methods, including formal verification, static analysis, and post-hoc explainability. We advocate exploring neuro-symbolic approaches for justification generation, where symbolic constraints guide model behavior during training and program semantics are enriched through neural representations, enabling automated consistency checks at inference time. |
| title | Position: Intelligent Coding Systems Should Write Programs with Justifications |
| topic | Software Engineering Computation and Language Machine Learning |
| url | https://arxiv.org/abs/2508.06017 |