# A Scalable Approach to Clustering Embedding Projections

arXiv: https://arxiv.org/abs/2504.07285 · HTML: https://mlsanity.com/paper/2504.07285/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 20 references that are in the map.
- First posted on arXiv 9 Apr 2025.

## Builds on (most cited first, 6 of 6)

- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction](https://mlsanity.com/paper/1802.03426/) (2018)
- [Visual Analytics in Deep Learning: An Interrogative Survey for the Next Frontiers](https://mlsanity.com/paper/1801.06889/) (2018)
- [Embedding Projector: Interactive Visualization and Interpretation of Embeddings](https://mlsanity.com/paper/1611.05469/) (2016)
- [WizMap: Scalable Interactive Visualization for Exploring Large Machine Learning Embeddings](https://mlsanity.com/paper/2306.09328/) (2023)
- [Visual Designs for Binned Aggregation of Multi-Class Scatterplots](https://mlsanity.com/paper/1810.02445/) (2018)

## Built on by (most cited first, 3 of 3)

- [TOUCAN: Synthesizing 1.5M Tool-Agentic Data from Real-World MCP Environments](https://mlsanity.com/paper/2510.01179/) (2025)
- [Embedding Atlas: Low-Friction, Interactive Embedding Visualization](https://mlsanity.com/paper/2505.06386/) (2025)
- [LangLasso: Interactive Cluster Descriptions through LLM Explanation](https://mlsanity.com/paper/2601.10458/) (2026)

## Abstract (arXiv)

Interactive visualization of embedding projections is a useful technique for understanding data and evaluating machine learning models. Labeling data within these visualizations is critical for interpretation, as labels provide an overview of the projection and guide user navigation. However, most methods for producing labels require clustering the points, which can be computationally expensive as the number of points grows. In this paper, we describe an efficient clustering approach using kernel density estimation in the projected 2D space instead of points. This algorithm can produce high-quality cluster regions from a 2D density map in a few hundred milliseconds, orders of magnitude faster than current approaches. We contribute the design of the algorithm, benchmarks, and applications that demonstrate the utility of the algorithm, including labeling and summarization.