Testing Deep Learning Library APIs via Cross-Framework Differential Fuzzing
Source
Bin Duan, Ruican Dong, Naipeng Dong, Dan Dongseong Kim, Guowei Yang
Author
Bin Duan, Ruican Dong, Naipeng Dong, Dan Dongseong Kim, Guowei Yang
Date
Why it matters
Defects that behave the same on CPU and GPU slip past the usual differential oracle; comparing equivalent operations across seven DL libraries catches APIs that quietly disagree on boundary and non-finite inputs.
articleHarnessing LLMs for Document-Guided Fuzzing of Python LibrariesBin Duan, Tarek Mahmud, Meiru Che, Yan Yan, Naipeng Dong, Dan Dongseong Kim, Guowei Yang
articleToolRobustBench: Stage-Wise Perturbation Evaluation and Failure Diagnosis for Tool-Calling AgentsYiShan Zheng, Yuan Wu, Yi Chang
articleThe Quality of Claude AI-authored Python Tests Is Not Weaker Than Human-authored TestsDouglas J. Leith