Intrinsic Concept Extraction Based on Compositional Interpretability

TL;DR

This paper introduces CI-ICE, a new unsupervised task for extracting composable, intrinsic concepts from a single image using diffusion models, and proposes HyperExpress to achieve hierarchical and relational concept disentanglement.

Contribution

The paper proposes HyperExpress, a novel method leveraging hyperbolic space and concept-wise optimization for compositional intrinsic concept extraction from images.

Findings

HyperExpress effectively disentangles hierarchical concepts.

The method preserves complex inter-concept relationships.

It achieves superior interpretability in concept extraction.

Abstract

Unsupervised Concept Extraction aims to extract concepts from a single image; however, existing methods suffer from the inability to extract composable intrinsic concepts. To address this, this paper introduces a new task called Compositional and Interpretable Intrinsic Concept Extraction (CI-ICE). The CI-ICE task aims to leverage diffusion-based text-to-image models to extract composable object-level and attribute-level concepts from a single image, such that the original concept can be reconstructed through the combination of these concepts. To achieve this goal, we propose a method called HyperExpress, which addresses the CI-ICE task through two core aspects. Specifically, first, we propose a concept learning approach that leverages the inherent hierarchical modeling capability of hyperbolic space to achieve accurate concept disentanglement while preserving the hierarchical structure and relational dependencies among concepts; second, we introduce a concept-wise optimization method that maps the concept embedding space to maintain complex inter-concept relationships while ensuring concept composability. Our method demonstrates outstanding performance in extracting compositionally interpretable intrinsic concepts from a single image.