Towards Trustworthy Explanation: On Causal Rationalization

by   Wenbo Zhang, et al.

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, we novelly leverage two causal desiderata, non-spuriousness and efficiency, into rationalization from the causal inference perspective. We formally define a series of probabilities of causation 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.


page 1

page 2

page 3

page 4


Relating Graph Neural Networks to Structural Causal Models

Causality can be described in terms of a structural causal model (SCM) t...

Causal Inference in Natural Language Processing: Estimation, Prediction, Interpretation and Beyond

A fundamental goal of scientific research is to learn about causal relat...

Causal Intervention for Subject-Deconfounded Facial Action Unit Recognition

Subject-invariant facial action unit (AU) recognition remains challengin...

Explaining The Behavior Of Black-Box Prediction Algorithms With Causal Learning

We propose to explain the behavior of black-box prediction methods (e.g....

Quantifying Sufficient Randomness for Causal Inference

Spurious association arises from covariance between propensity for the t...

Algorithmic Causal Effect Identification with causaleffect

Our evolution as a species made a huge step forward when we understood t...

Causal Inference via Nonlinear Variable Decorrelation for Healthcare Applications

Causal inference and model interpretability research are gaining increas...

Please sign up or login with your details

Forgot password? Click here to reset