Data-driven Test Generation for Fuzzing AI Compiler
Fuente:
arXiv
Saved in:
| Main Author: | |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912845597245440 |
|---|---|
| author | Shen, Qingchao |
| author_facet | Shen, Qingchao |
| contents | Artificial Intelligence (AI) compilers are critical for efficiently deploying AI models across diverse hardware platforms. However, they remain prone to bugs that can compromise both compiler reliability and model correctness. Thus, ensuring the quality of AI compilers is crucial. In this work, we present a unified data-driven testing framework that systematically addresses stage-specific challenges in AI compilers. Specifically, OPERA migrates tests for AI libraries to test various operator conversion logic in the model loading stage. OATest synthesizes diverse optimization-aware computational graphs for testing high-level optimizations. HARMONY generates and mutates diverse low-level IR seeds to generate hardware-optimization-aware tests for testing low-level optimizations. Together, these techniques provide a comprehensive, stage-aware framework that enhances testing coverage and effectiveness, detecting 266 previously unknown bugs in four widely used AI compilers. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_17450 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Data-driven Test Generation for Fuzzing AI Compiler Shen, Qingchao Software Engineering Artificial Intelligence (AI) compilers are critical for efficiently deploying AI models across diverse hardware platforms. However, they remain prone to bugs that can compromise both compiler reliability and model correctness. Thus, ensuring the quality of AI compilers is crucial. In this work, we present a unified data-driven testing framework that systematically addresses stage-specific challenges in AI compilers. Specifically, OPERA migrates tests for AI libraries to test various operator conversion logic in the model loading stage. OATest synthesizes diverse optimization-aware computational graphs for testing high-level optimizations. HARMONY generates and mutates diverse low-level IR seeds to generate hardware-optimization-aware tests for testing low-level optimizations. Together, these techniques provide a comprehensive, stage-aware framework that enhances testing coverage and effectiveness, detecting 266 previously unknown bugs in four widely used AI compilers. |
| title | Data-driven Test Generation for Fuzzing AI Compiler |
| topic | Software Engineering |
| url | https://arxiv.org/abs/2601.17450 |