A1301
Title: Towards trustworthy explanation: On causal rationalization
Authors: Hengrui Cai - University of California Irvine (United States) [presenting]
Abstract: With recent advances in natural language processing, rationalization becomes an essential self-explaining diagram to disentangle the black box by selecting a subset of input texts to account for the major variation in prediction. Yet, existing association-based approaches on rationalization cannot identify true rationales when two or more snippets are highly inter-correlated and thus provide a similar contribution to prediction accuracy, so-called spuriousness. To address this limitation, two causal desiderata, non-spuriousness and efficiency, are novelly leveraged into rationalization from the causal inference perspective. A series of probabilities of causation is formally defined based on a newly proposed structural causal model of rationalization, with its theoretical identification established as the main component of learning necessary and sufficient rationales. The superior performance of the proposed causal rationalization is demonstrated on real-world review and medical datasets with extensive experiments compared to state-of-the-art methods.
