Papers by Aron Culotta
Identifying Spurious Correlations for Robust Text Classification (2020.findings-emnlp)
Copied to clipboard
| Challenge: | Text classifiers often rely on spurious correlations to predict positive reviews . term Spielberg does not cause the review to be positive, so it does not affect the classification accuracy. |
| Approach: | They propose a method to distinguish spurious and genuine correlations in text classification using treatment effect estimators. |
| Outcome: | The proposed method works well even with limited training examples and is possible to transport the word classifier to new domains. |
Using Text-Based Causal Inference to Disentangle Factors Influencing Online Review Ratings (2025.naacl-long)
Copied to clipboard
| Challenge: | Existing methods to analyze online reviews for aspects of quality are limited . authors propose a method to disentangle the impact of each aspect on overall perception . |
| Approach: | They propose a method to disentangle the effect of each aspect on overall perception . they use textual mentions in reviews as proxies for real-world attributes . |
| Outcome: | The proposed method improves on real-world reviews of U.S. K-12 schools. |