# Text-to-Image Models for Counterfactual Explanations: a Black-Box Approach

arXiv: https://arxiv.org/abs/2309.07944 · HTML: https://mlsanity.com/paper/2309.07944/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Builds on 48 of its 60 references that are in the map.
- First posted on arXiv 14 Sep 2023.

## Builds on (most cited first, 30 of 48)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Grad-CAM: Visual Explanations from Deep Networks via Gradient-based Localization](https://mlsanity.com/paper/1610.02391/) (2016)
- [Densely Connected Convolutional Networks](https://mlsanity.com/paper/1608.06993/) (2016)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [DreamBooth: Fine Tuning Text-to-Image Diffusion Models for Subject-Driven Generation](https://mlsanity.com/paper/2208.12242/) (2022)
- [Taming Transformers for High-Resolution Image Synthesis](https://mlsanity.com/paper/2012.09841/) (2020)
- [An Image is Worth One Word: Personalizing Text-to-Image Generation using Textual Inversion](https://mlsanity.com/paper/2208.01618/) (2022)
- [T2I-Adapter: Learning Adapters to Dig out More Controllable Ability for Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.08453/) (2023)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [BDD100K: A Diverse Driving Dataset for Heterogeneous Multitask Learning](https://mlsanity.com/paper/1805.04687/) (2018)
- [Training Generative Adversarial Networks with Limited Data](https://mlsanity.com/paper/2006.06676/) (2020)
- [Null-text Inversion for Editing Real Images using Guided Diffusion Models](https://mlsanity.com/paper/2211.09794/) (2022)
- [Counterfactual Explanations without Opening the Black Box: Automated Decisions and the GDPR](https://mlsanity.com/paper/1711.00399/) (2017)
- [Deep Leakage from Gradients](https://mlsanity.com/paper/1906.08935/) (2019)
- [Grad-CAM++: Improved Visual Explanations for Deep Convolutional Networks](https://mlsanity.com/paper/1710.11063/) (2017)
- [RISE: Randomized Input Sampling for Explanation of Black-box Models](https://mlsanity.com/paper/1806.07421/) (2018)
- [This Looks Like That: Deep Learning for Interpretable Image Recognition](https://mlsanity.com/paper/1806.10574/) (2018)
- [MaskGAN: Towards Diverse and Interactive Facial Image Manipulation](https://mlsanity.com/paper/1907.11922/) (2019)
- [Zero-shot Image-to-Image Translation](https://mlsanity.com/paper/2302.03027/) (2023)
- [EDICT: Exact Diffusion Inversion via Coupled Transformations](https://mlsanity.com/paper/2211.12446/) (2022)
- [CRAFT: Concept Recursive Activation FacTorization for Explainability](https://mlsanity.com/paper/2211.10154/) (2022)
- [Interpretable Convolutional Neural Networks](https://mlsanity.com/paper/1710.00935/) (2017)
- [Deformable ProtoPNet: An Interpretable Image Classifier Using Deformable Prototypes](https://mlsanity.com/paper/2111.15000/) (2021)
- [Prompt Tuning Inversion for Text-Driven Image Editing Using Diffusion Models](https://mlsanity.com/paper/2305.04441/) (2023)
- [Diffusion Visual Counterfactual Explanations](https://mlsanity.com/paper/2210.11841/) (2022)

## Built on by (most cited first, 28 of 28)

- [Global Counterfactual Directions](https://mlsanity.com/paper/2404.12488/) (2024)
- [Rethinking Visual Counterfactual Explanations Through Region Constraint](https://mlsanity.com/paper/2410.12591/) (2024)
- [Towards Desiderata-Driven Design of Visual Counterfactual Explainers](https://mlsanity.com/paper/2506.14698/) (2025)
- [Benchmarking Segmentation Models with Mask-Preserved Attribute Editing](https://mlsanity.com/paper/2403.01231/) (2024)
- [Diffusion Counterfactuals for Image Regressors](https://mlsanity.com/paper/2503.20595/) (2025)
- [V-CECE: Visual Counterfactual Explanations via Conceptual Edits](https://mlsanity.com/paper/2509.16567/) (2025)
- [Graph Diffusion Counterfactual Explanation](https://mlsanity.com/paper/2511.16287/) (2025)
- [Visual Disentangled Diffusion Autoencoders: Scalable Counterfactual Generation for Foundation Models](https://mlsanity.com/paper/2601.21851/) (2026)
- [Learning Common and Salient Generative Factors Between Two Image Datasets](https://mlsanity.com/paper/2512.12800/) (2025)
- [SCE-LITE-HQ: Smooth visual counterfactual explanations with generative foundation models](https://mlsanity.com/paper/2603.17048/) (2026)
- [GIFT: A Framework Towards Global Interpretable Faithful Textual Explanations of Vision Classifiers](https://mlsanity.com/paper/2411.15605/) (2024)
- [Photorealistic Inpainting for Perturbation-based Explanations in Ecological Monitoring](https://mlsanity.com/paper/2510.03317/) (2025)
- [Auditing Sybil: Explaining Deep Lung Cancer Risk Prediction Through Generative Interventional Attributions](https://mlsanity.com/paper/2602.02560/) (2026)
- [MaskDiME: Adaptive Masked Diffusion for Precise and Efficient Visual Counterfactual Explanations](https://mlsanity.com/paper/2602.18792/) (2026)
- [Benchmarking Semantic Segmentation Models via Appearance and Geometry Attribute Editing](https://mlsanity.com/paper/2603.01535/) (2026)
- [Teaching Humans Subtle Differences with DIFFusion](https://mlsanity.com/paper/2504.08046/) (2025)
- [Back to the Feature: Explaining Video Classifiers with Video Counterfactual Explanations](https://mlsanity.com/paper/2511.20295/) (2025)
- [Faithful, Sufficient and Understandable: Rethinking Graph Counterfactual Explanations via Discrete Diffusion Inversion](https://mlsanity.com/paper/2608.12083/) (2026)
- [DiffEx: Explaining a Classifier with Diffusion Models to Identify Microscopic Cellular Variations](https://mlsanity.com/paper/2502.09663/) (2025)
- [Unsupervised Structural-Counterfactual Generation under Domain Shift](https://mlsanity.com/paper/2502.12013/) (2025)
- [Counterfactual Explanation for Multivariate Time Series Forecasting with Exogenous Variables](https://mlsanity.com/paper/2511.06906/) (2025)
- [Bot Meets Shortcut: How Can LLMs Aid in Handling Unknown Invariance OOD Scenarios?](https://mlsanity.com/paper/2511.08455/) (2025)
- [Cross-modal Counterfactual Explanations: Uncovering Decision Factors and Dataset Biases in Subjective Classification](https://mlsanity.com/paper/2512.18864/) (2025)
- [FiRe: Fixed-Noise Refinement for Visual Counterfactual Explanations](https://mlsanity.com/paper/2608.08664/) (2026)
- [Counterfactual Contrastive Analysis](https://mlsanity.com/paper/2608.19032/) (2026)
- [MedDiME: Efficient Latent Diffusion with Adaptive Masking for Medical Counterfactual Generation](https://mlsanity.com/paper/2609.15647/) (2026)
- [Targeted Visual Counterfactual Explanations for Contrastive Vision-Language Model](https://mlsanity.com/paper/2609.37638/) (2026)
- [Towards Fast and Disentangled Counterfactuals for Visual Foundation Models](https://mlsanity.com/paper/2610.00895/) (2026)

## Abstract (arXiv)

This paper addresses the challenge of generating Counterfactual Explanations (CEs), involving the identification and modification of the fewest necessary features to alter a classifier's prediction for a given image. Our proposed method, Text-to-Image Models for Counterfactual Explanations (TIME), is a black-box counterfactual technique based on distillation. Unlike previous methods, this approach requires solely the image and its prediction, omitting the need for the classifier's structure, parameters, or gradients. Before generating the counterfactuals, TIME introduces two distinct biases into Stable Diffusion in the form of textual embeddings: the context bias, associated with the image's structure, and the class bias, linked to class-specific features learned by the target classifier. After learning these biases, we find the optimal latent code applying the classifier's predicted class token and regenerate the image using the target embedding as conditioning, producing the counterfactual explanation. Extensive empirical studies validate that TIME can generate explanations of comparable effectiveness even when operating within a black-box setting.