# Interpretable Diffusion Models with B-cos Networks

arXiv: https://arxiv.org/abs/2507.03846 · HTML: https://mlsanity.com/paper/2507.03846/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Image generation, text-to-image and editing, one of 14 directions in Diffusion models.
- Builds on 21 of its 25 references that are in the map.
- First posted on arXiv 5 Jul 2025.

## Placed in

- Diffusion models / [Image generation, text-to-image and editing](https://mlsanity.com/diffusion/image-generation-text-to-image-and-editing/)

## Builds on (most cited first, 21 of 21)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [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)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- ["Why Should I Trust You?": Explaining the Predictions of Any Classifier](https://mlsanity.com/paper/1602.04938/) (2016)
- [LAION-5B: An open large-scale dataset for training next generation image-text models](https://mlsanity.com/paper/2210.08402/) (2022)
- [CLIPScore: A Reference-free Evaluation Metric for Image Captioning](https://mlsanity.com/paper/2104.08718/) (2021)
- [What the DAAM: Interpreting Stable Diffusion Using Cross Attention](https://mlsanity.com/paper/2210.04885/) (2022)
- [Full-Gradient Representation for Neural Network Visualization](https://mlsanity.com/paper/1905.00780/) (2019)
- [Diffuse, Attend, and Segment: Unsupervised Zero-Shot Segmentation using Stable Diffusion](https://mlsanity.com/paper/2308.12469/) (2023)
- [B-cos Networks: Alignment is All We Need for Interpretability](https://mlsanity.com/paper/2205.10268/) (2022)
- [Diffusion Model is Secretly a Training-free Open Vocabulary Semantic Segmenter](https://mlsanity.com/paper/2309.02773/) (2023)
- [Self-Discovering Interpretable Diffusion Latent Directions for Responsible Text-to-Image Generation](https://mlsanity.com/paper/2311.17216/) (2023)
- [Prompt-Free Diffusion: Taking "Text" out of Text-to-Image Diffusion Models](https://mlsanity.com/paper/2305.16223/) (2023)
- [Interpretable Diffusion via Information Decomposition](https://mlsanity.com/paper/2310.07972/) (2023)
- [B-cos Alignment for Inherently Interpretable CNNs and Vision Transformers](https://mlsanity.com/paper/2306.10898/) (2023)
- [On Discrete Prompt Optimization for Diffusion Models](https://mlsanity.com/paper/2407.01606/) (2024)
- [B-cosification: Transforming Deep Neural Networks to be Inherently Interpretable](https://mlsanity.com/paper/2411.00715/) (2024)
- [B-Cos Aligned Transformers Learn Human-Interpretable Features](https://mlsanity.com/paper/2401.08868/) (2024)

## Abstract (arXiv)

Text-to-image diffusion models generate images by iteratively denoising random noise, conditioned on a prompt. While these models have enabled impressive progress in image generation, they often fail to accurately reflect all semantic information described in the prompt -- failures that are difficult to detect automatically. In this work, we introduce a diffusion model architecture built with B-cos modules that offers inherent interpretability. Our approach provides insight into how individual prompt tokens affect the generated image by producing explanations that highlight the pixel regions influenced by each token. We demonstrate that B-cos diffusion models can produce high-quality images while providing meaningful insights into prompt-image alignment.