Towards Achieving Concept Completeness for Unsupervised Textual Concept Bottleneck Models

Milan Bhan, Yann Choho, Pierre Moreau, Jean-Noël Vittaut, Nicolas Chesneau, Marie-Jeanne Lesot

Abstract: Textual Concept Bottleneck Models (TCBMs) are interpretable-by-design models for text classification that predict a set of salient concepts before making the final prediction. This paper proposes Complete Textual Concept Bottleneck Model (CT-CBM), a novel TCBM generator building concept labels in a fully unsupervised manner using a small language model, eliminating both the need for predefined human labeled concepts and LLM annotations. CT-CBM iteratively targets and adds important and identifiable concepts in the bottleneck layer to create a complete concept basis. CT-CBM achieves striking results against competitors in terms of concept basis completeness and concept detection accuracy, offering a promising solution to reliably enhance interpretability of NLP classifiers.

External IDs:dblp:journals/corr/abs-2502-11100