Search papers, labs, and topics across Lattice.
This paper introduces Xamt, a novel cross-framework differential fuzzing technique designed to test deep learning library APIs across multiple frameworks. By validating API correspondences through execution and behavioral checks, Xamt identifies discrepancies that traditional methods might overlook, revealing critical reliability issues. The approach successfully constructs 676 execution-validated API groups and uncovers 72 significant discrepancies, leading to 25 confirmed developer reports, including 23 fixes.
Xamt uncovers critical API discrepancies across deep learning libraries, revealing that traditional testing methods may miss up to 72 significant issues.
Deep learning libraries underpin many safety- and reliability-critical applications, yet existing API-level testing techniques often rely on intra-library properties or CPU--GPU differential oracles and may miss defects that behave consistently across hardware backends. We present Xamt, a cross-framework differential fuzzing approach for deep learning library APIs. Xamt constructs and tests execution-validated groups of APIs intended to implement equivalent operations across seven libraries. It uses explicit API aliases and parameter-role normalization to construct candidate correspondences and validates them through pairwise execution and a group-level behavioral check on canonical ordinary inputs. The resulting groups are explored using variance-guided differential fuzzing with ordinary, boundary, and non-finite inputs. Crash and inconsistency oracles flag executions exhibiting abnormal termination or inconsistent outputs for subsequent reproduction and analysis. Across the seven libraries, Xamt constructs 676 execution-validated groups containing 2,563 matched APIs. Among these, Xamt identifies 72 independently reproduced discrepancy cases, including 4 crash cases and 68 output inconsistencies. Among the 72 developer reports, 25 have been confirmed, including 23 that have been fixed.
