# How Far Does a Shared Linear Map Go? Probing Feature-Space Manipulability for Image Editing

arXiv: https://arxiv.org/abs/2605.11203 · HTML: https://mlsanity.com/paper/2605.11203/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 31 references that are in the map.
- First posted on arXiv 11 May 2026.

## Builds on (most cited first, 26 of 26)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [Benchmarking Neural Network Robustness to Common Corruptions and Perturbations](https://mlsanity.com/paper/1903.12261/) (2019)
- [Sparse Autoencoders Find Highly Interpretable Features in Language Models](https://mlsanity.com/paper/2309.08600/) (2023)
- [LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop](https://mlsanity.com/paper/1506.03365/) (2015)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [Qwen-Image Technical Report](https://mlsanity.com/paper/2508.02324/) (2025)
- [Mechanistic Interpretability for AI Safety -- A Review](https://mlsanity.com/paper/2404.14082/) (2024)
- [Understanding image representations by measuring their equivariance and equivalence](https://mlsanity.com/paper/1411.5908/) (2014)
- [Revisiting Model Stitching to Compare Neural Representations](https://mlsanity.com/paper/2106.07682/) (2021)
- [Relative representations enable zero-shot latent space communication](https://mlsanity.com/paper/2209.15430/) (2022)
- [CRAFT: Concept Recursive Activation FacTorization for Explainability](https://mlsanity.com/paper/2211.10154/) (2022)
- [The Geometry of Categorical and Hierarchical Concepts in Large Language Models](https://mlsanity.com/paper/2406.01506/) (2024)
- [A Practical Review of Mechanistic Interpretability for Transformer-Based Language Models](https://mlsanity.com/paper/2407.02646/) (2024)
- [Implicit Semantic Data Augmentation for Deep Networks](https://mlsanity.com/paper/1909.12220/) (2019)
- [A Comprehensive Study on Robustness of Image Classification Models: Benchmarking and Rethinking](https://mlsanity.com/paper/2302.14301/) (2023)
- [Archetypal SAE: Adaptive and Stable Dictionary Learning for Concept Extraction in Large Vision Models](https://mlsanity.com/paper/2502.12892/) (2025)
- [Battle of the Backbones: A Large-Scale Comparison of Pretrained Models across Computer Vision Tasks](https://mlsanity.com/paper/2310.19909/) (2023)
- [High Fidelity Visualization of What Your Self-Supervised Representation Knows About](https://mlsanity.com/paper/2112.09164/) (2021)
- [Learning by Reconstruction Produces Uninformative Features For Perception](https://mlsanity.com/paper/2402.11337/) (2024)
- [Multi-dimensional concept discovery (MCD): A unifying framework with completeness guarantees](https://mlsanity.com/paper/2301.11911/) (2023)
- [Linear Spaces of Meanings: Compositional Structures in Vision-Language Models](https://mlsanity.com/paper/2302.14383/) (2023)
- [Into the Rabbit Hull: From Task-Relevant Concepts in DINO to Minkowski Geometry](https://mlsanity.com/paper/2510.08638/) (2025)
- [Do Sparse Autoencoders Capture Concept Manifolds?](https://mlsanity.com/paper/2604.28119/) (2026)
- [Beyond Scalars: Concept-Based Alignment Analysis in Vision Transformers](https://mlsanity.com/paper/2412.06639/) (2024)
- [FeatInv: Spatially resolved mapping from feature space to input space using conditional diffusion models](https://mlsanity.com/paper/2505.21032/) (2025)

## Abstract (arXiv)

Understanding how image-space transformations manifest in a model's internal representations is a longstanding goal in representation analysis. Prior work has shown that geometric transformations can often be captured by learned linear operators between feature maps, but it remains unclear whether this extends to photometric, local, and semantically defined edits. We train probes of increasing capacity from a spatially shared linear map to nonlinear per-vector, receptive-field, and global transformer models to predict feature-space changes induced by geometric transforms, photometric edits, occlusions, and diffusion-generated semantic edits. Across ConvNeXt, SwinV2, and DINOv3, a single shared linear map often predicts held-out manipulation outcomes nearly as well as substantially more expressive probes for the supervised backbones, with sufficiency generally increasing with depth; this pattern is less consistent for DINOv3. These results suggest that a simple spatially shared linear operator is often sufficient to represent diverse image manipulations, while its leading singular components capture semantic content and higher-rank components primarily refine image details. We frame these findings as predictive representational sufficiency rather than evidence of intrinsic linear feature-space geometry.