Go to AAAI 2025 Workshop AICT homepageRevealing the Limitations of Exploiting Causal Effects to Resolve Linguistic Spurious Correlations
Keywords: causally related words, necessity, sufficiency, robust text classifier
TL;DR: This paper proposes a robust text classification approach using PN and PS to distinguish causally related words from spuriously correlated words.
Abstract: Identifying causal relationships rather than spurious correlations between words and class labels plays a crucial role in building robust text classifiers. Previous studies proposed using causal effect to distinguish words that are causally related to the sentiment, and then building robust text classifiers using words with high causal effects. However, we find that when a sentence has multiple causally related words simultaneously, the magnitude of causal effects will be significantly reduced, which limits the applicability of previous causal effect-based methods in distinguishing causally related words from spurious correlated ones. To fill this gap, in this paper, we introduce both the probability of necessity (PN) and probability of sufficiency (PS), aiming to answer the counterfactual question that `if a sentence has a certain sentiment in the presence/absence of a word, would the sentiment change in the absence/presence of that word?'. Specifically, we first derive the identifiability of PN and PS under different sentiment monotonicities, and calibrate the estimation of PN and PS via the estimated average treatment effect, finally the robust text classifier is built by removing a certain percentage of words with the lowest estimated PN and PS. Extensive experiments are conducted on public datasets to validate the effectiveness of our method.
Submission Number: 25
Loading