PE: A Poincare Explanation Method for Fast Text Hierarchy Generation
This work addresses the need for interpretability in NLP models by providing a faster method for hierarchical attribution, though it is incremental as it builds on prior work in hyperbolic spaces and Shapley scores.
The paper tackles the problem of generating text hierarchies for explaining deep NLP models by introducing Poincare Explanation (PE), which uses hyperbolic spaces to model feature interactions efficiently, resulting in a time-efficient method that outperforms existing approaches.
The black-box nature of deep learning models in NLP hinders their widespread application. The research focus has shifted to Hierarchical Attribution (HA) for its ability to model feature interactions. Recent works model non-contiguous combinations with a time-costly greedy search in Eculidean spaces, neglecting underlying linguistic information in feature representations. In this work, we introduce a novel method, namely Poincare Explanation (PE), for modeling feature interactions with hyperbolic spaces in a time efficient manner. Specifically, we take building text hierarchies as finding spanning trees in hyperbolic spaces. First we project the embeddings into hyperbolic spaces to elicit inherit semantic and syntax hierarchical structures. Then we propose a simple yet effective strategy to calculate Shapley score. Finally we build the the hierarchy with proving the constructing process in the projected space could be viewed as building a minimum spanning tree and introduce a time efficient building algorithm. Experimental results demonstrate the effectiveness of our approach.