Using a corpus of 5542 fault-injected traces from 38 DL programs, the study finds a 0.19 balanced accuracy gap in fault diagnosis between within-program and cross-program evaluation caused by program-specific feature structures.
Ruishi Chen, Victor R Lee, Annie Camey Kuo, Denise Clark Pope, and Sarah Miles
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
verdicts
UNVERDICTED 4representative citing papers
Generative AI reduced study time on AI-susceptible math problems by 9-31% across grade levels and produced a 25% decline in retention odds on proctored assessments.
VarQEC uses a distinguishability loss as a machine-learning objective to variationally discover resource-efficient encoding circuits optimized for given noise models.
Using distribution regression on Consumption Expenditure Interview Survey data, the study decomposes the 2018-2022 decline in consumption inequality into contributions from conditional consumption distributions, rising asset holdings, and household characteristics for male-headed households.
citing papers explorer
-
Evaluation-Strategy Gap in Fault Diagnosis of Deep Learning Programs
Using a corpus of 5542 fault-injected traces from 38 DL programs, the study finds a 0.19 balanced accuracy gap in fault diagnosis between within-program and cross-program evaluation caused by program-specific feature structures.