# Intrinsic Concept Extraction Based on Compositional Interpretability

Source: [Intrinsic Concept Extraction Based on Compositional Interpretability](<https://www.sota2.com/research/paper/69b39a210636e5bf4afbcb29>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-14T05:00:04.234Z

Authors: Hanyu Shi, Hong Tao, Guoheng Huang, Jianbin Jiang, Xuhang Chen, Chi-Man Pun, Shanhu Wang, Pan Pan

Published: 2026-03-12T10:56:05.000Z

## 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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Intrinsic Concept Extraction on ICBench D1](<https://www.sota2.com/research/sota/intrinsic-concept-extraction-on-icbench-d1.md>): Intrinsic Concept Extraction; ICBench D1; SIMT-T Score (Object): 28.

- [Unsupervised Concept Extraction on UCEBench](<https://www.sota2.com/research/sota/unsupervised-concept-extraction-on-ucebench.md>): Unsupervised Concept Extraction; UCEBench; SIMI: 69.9.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.11795>)

[PDF](<https://arxiv.org/pdf/2603.11795.pdf>)