Interpretable-by-Design Text Understanding with Iteratively Generated Concept Bottleneck
This addresses the problem of interpretability in text classification for high-stakes domains, offering a novel framework with minimal performance tradeoffs.
The authors tackled the lack of interpretability in text classification for high-stakes domains by proposing Text Bottleneck Models (TBM), which predict categorical values for sparse concepts and use a linear layer for final predictions, achieving performance rivaling few-shot GPT-4 and finetuned DeBERTa on 12 datasets but falling short against finetuned GPT-3.5.
Black-box deep neural networks excel in text classification, yet their application in high-stakes domains is hindered by their lack of interpretability. To address this, we propose Text Bottleneck Models (TBM), an intrinsically interpretable text classification framework that offers both global and local explanations. Rather than directly predicting the output label, TBM predicts categorical values for a sparse set of salient concepts and uses a linear layer over those concept values to produce the final prediction. These concepts can be automatically discovered and measured by a Large Language Model (LLM) without the need for human curation. Experiments on 12 diverse text understanding datasets demonstrate that TBM can rival the performance of black-box baselines such as few-shot GPT-4 and finetuned DeBERTa while falling short against finetuned GPT-3.5. Comprehensive human evaluation validates that TBM can generate high-quality concepts relevant to the task, and the concept measurement aligns well with human judgments, suggesting that the predictions made by TBMs are interpretable. Overall, our findings suggest that TBM is a promising new framework that enhances interpretability with minimal performance tradeoffs.