# Naming the Concepts Classifiers Rely On: Language-Anchored Decomposition for Faithful Explanation

arXiv: https://arxiv.org/abs/2607.07264 · HTML: https://mlsanity.com/paper/2607.07264/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 27 references that are in the map.
- First posted on arXiv 8 Jul 2026.

## Builds on (most cited first, 22 of 22)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps](https://mlsanity.com/paper/1312.6034/) (2013)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [SmoothGrad: removing noise by adding noise](https://mlsanity.com/paper/1706.03825/) (2017)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [RISE: Randomized Input Sampling for Explanation of Black-box Models](https://mlsanity.com/paper/1806.07421/) (2018)
- [Network Dissection: Quantifying Interpretability of Deep Visual Representations](https://mlsanity.com/paper/1704.05796/) (2017)
- [Label-Free Concept Bottleneck Models](https://mlsanity.com/paper/2304.06129/) (2023)
- [What does CLIP know about a red circle? Visual prompt engineering for VLMs](https://mlsanity.com/paper/2304.06712/) (2023)
- [CRAFT: Concept Recursive Activation FacTorization for Explainability](https://mlsanity.com/paper/2211.10154/) (2022)
- [CLIP-Dissect: Automatic Description of Neuron Representations in Deep Vision Networks](https://mlsanity.com/paper/2204.10965/) (2022)
- [Natural Language Descriptions of Deep Visual Features](https://mlsanity.com/paper/2201.11114/) (2022)
- [Invertible Concept-based Explanations for CNN Models with Non-negative Concept Activation Vectors](https://mlsanity.com/paper/2006.15417/) (2020)
- [What I Cannot Predict, I Do Not Understand: A Human-Centered Evaluation Framework for Explainability Methods](https://mlsanity.com/paper/2112.04417/) (2021)
- [The effectiveness of feature attribution methods and its correlation with automatic evaluation scores](https://mlsanity.com/paper/2105.14944/) (2021)
- [General Pitfalls of Model-Agnostic Interpretation Methods for Machine Learning Models](https://mlsanity.com/paper/2007.04131/) (2020)
- [Interpretability Beyond Classification Output: Semantic Bottleneck Networks](https://mlsanity.com/paper/1907.10882/) (2019)
- [Learning Bottleneck Concepts in Image Classification](https://mlsanity.com/paper/2304.10131/) (2023)
- [Show and Tell: Visually Explainable Deep Neural Nets via Spatially-Aware Concept Bottleneck Models](https://mlsanity.com/paper/2502.20134/) (2025)
- [Explaining Deep Convolutional Neural Networks via Latent Visual-Semantic Filter Attention](https://mlsanity.com/paper/2204.04601/) (2022)
- [FACE: Faithful Automatic Concept Extraction](https://mlsanity.com/paper/2510.11675/) (2025)

## Abstract (arXiv)

Deep neural networks are widely deployed in high-stakes visual applications where interpretability is critical, yet existing explanations face a trade-off: post-hoc concept methods recover factors that are faithful to a model's behavior but unnamed, while naming and by-design methods attach human-readable concepts only by retraining or altering the classifier. We propose Language-Anchored Decomposition (LAD), a post-hoc framework that delivers concepts which are simultaneously named, faithful, and obtained without modifying the model. For each class, a large language model proposes a concept vocabulary that CLIP-based similarity maps localize across image regions. Inverting standard non-negative matrix factorization, LAD fixes these language-grounded maps as the coefficient matrix and learns only a concept basis that reconstructs the frozen encoder's activations, so naming becomes a structural constraint and the model's own feature geometry determines which concepts are retained. Removing this anchor preserves accuracy but collapses attribution faithfulness. Across natural-image, scene, and medical-imaging benchmarks, LAD produces spatially precise explanations that are decision-relevant under both concept insertion and deletion, while uniquely providing stable, human-interpretable concept names.