# Interpreting the structure of multi-object representations in vision encoders

arXiv: https://arxiv.org/abs/2406.09067 · HTML: https://mlsanity.com/paper/2406.09067/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 30 of its 42 references that are in the map.
- First posted on arXiv 13 Jun 2024.

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

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [DINOv2: Learning Robust Visual Features without Supervision](https://mlsanity.com/paper/2304.07193/) (2023)
- [Emerging Properties in Self-Supervised Vision Transformers](https://mlsanity.com/paper/2104.14294/) (2021)
- [Neural Discrete Representation Learning](https://mlsanity.com/paper/1711.00937/) (2017)
- [BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation](https://mlsanity.com/paper/2201.12086/) (2022)
- [Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations](https://mlsanity.com/paper/1602.07332/) (2016)
- [VQA: Visual Question Answering](https://mlsanity.com/paper/1505.00468/) (2015)
- [CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning](https://mlsanity.com/paper/1612.06890/) (2016)
- [Understanding intermediate layers using linear classifier probes](https://mlsanity.com/paper/1610.01644/) (2016)
- [Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges](https://mlsanity.com/paper/2104.13478/) (2021)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [Object-Centric Learning with Slot Attention](https://mlsanity.com/paper/2006.15055/) (2020)
- [Do Vision Transformers See Like Convolutional Neural Networks?](https://mlsanity.com/paper/2108.08810/) (2021)
- [FLAVA: A Foundational Language And Vision Alignment Model](https://mlsanity.com/paper/2112.04482/) (2021)
- [Designing and Interpreting Probes with Control Tasks](https://mlsanity.com/paper/1909.03368/) (2019)
- [CLIPort: What and Where Pathways for Robotic Manipulation](https://mlsanity.com/paper/2109.12098/) (2021)
- [When and why vision-language models behave like bags-of-words, and what to do about it?](https://mlsanity.com/paper/2210.01936/) (2022)
- [Mask DINO: Towards A Unified Transformer-based Framework for Object Detection and Segmentation](https://mlsanity.com/paper/2206.02777/) (2022)
- [Neural-Symbolic VQA: Disentangling Reasoning from Vision and Language Understanding](https://mlsanity.com/paper/1810.02338/) (2018)
- [On the Relationship between Self-Attention and Convolutional Layers](https://mlsanity.com/paper/1911.03584/) (2019)
- [On the Binding Problem in Artificial Neural Networks](https://mlsanity.com/paper/2012.05208/) (2020)
- [Does CLIP Bind Concepts? Probing Compositionality in Large Image Models](https://mlsanity.com/paper/2212.10537/) (2022)
- [Behind the Scene: Revealing the Secrets of Pre-trained Vision-and-Language Models](https://mlsanity.com/paper/2005.07310/) (2020)
- [VL-InterpreT: An Interactive Visualization Tool for Interpreting Vision-Language Transformers](https://mlsanity.com/paper/2203.17247/) (2022)
- [Break It Down: Evidence for Structural Compositionality in Neural Networks](https://mlsanity.com/paper/2301.10884/) (2023)
- [Do Vision-Language Pretrained Models Learn Composable Primitive Concepts?](https://mlsanity.com/paper/2203.17271/) (2022)
- [Unit Testing for Concepts in Neural Networks](https://mlsanity.com/paper/2208.10244/) (2022)
- [Discrete Key-Value Bottleneck](https://mlsanity.com/paper/2207.11240/) (2022)
- [DINOv2 based Self Supervised Learning For Few Shot Medical Image Segmentation](https://mlsanity.com/paper/2403.03273/) (2024)

## Built on by (most cited first, 1 of 1)

- [Feature Evolution and Migration during Vision Transformer Training](https://mlsanity.com/paper/2608.20134/) (2026)

## Abstract (arXiv)

In this work, we interpret the representations of multi-object scenes in vision encoders through the lens of structured representations. Structured representations allow modeling of individual objects distinctly and their flexible use based on the task context for both scene-level and object-specific tasks. These capabilities play a central role in human reasoning and generalization, allowing us to abstract away irrelevant details and focus on relevant information in a compact and usable form. We define structured representations as those that adhere to two specific properties: binding specific object information into discrete representation units and segregating object representations into separate sets of tokens to minimize cross-object entanglement. Based on these properties, we evaluated and compared image encoders pre-trained on classification (ViT), large vision-language models (CLIP, BLIP, FLAVA), and self-supervised methods (DINO, DINOv2). We examine the token representations by creating object-decoding tasks that measure the ability of specific tokens to capture individual objects in multi-object scenes from the COCO dataset. This analysis provides insights into how object-wise representations are distributed across tokens and layers within these vision encoders. Our findings highlight significant differences in the representation of objects depending on their relevance to the pre-training objective, with this effect particularly pronounced in the CLS token (often used for downstream tasks). Meanwhile, networks and layers that exhibit more structured representations retain better information about individual objects. To guide practical applications, we propose formal measures to quantify the two properties of structured representations, aiding in selecting and adapting vision encoders for downstream tasks.