Less is Better: Recovering Intended-Feature Subspace to Robustify NLU Models. (arXiv:2209.07879v1 [cs.CL])

Datasets with significant proportions of bias present threats for training a
trustworthy model on NLU tasks. Despite yielding great progress, current
debiasing methods impose excessive reliance on the knowledge of bias
attributes. Definition of the attributes, however, is elusive and varies across
different datasets. Furthermore, leveraging these attributes at input level to
bias mitigation may leave a gap between intrinsic properties and the underlying
decision rule. To narrow down this gap and liberate the supervision on bias, we
suggest extending …

arxiv feature nlu