# Towards a Learning Theory of Representation Alignment

arXiv: https://arxiv.org/abs/2502.14047 · HTML: https://mlsanity.com/paper/2502.14047/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Feb 2025.

## Built on by (most cited first, 5 of 5)

- [Text-to-3D by Stitching a Multi-view Reconstruction Network to a Video Generator](https://mlsanity.com/paper/2510.13454/) (2025)
- [Cross-Model Semantics in Representation Learning](https://mlsanity.com/paper/2508.03649/) (2025)
- [Contrastive Time Series Forecasting with Anomalies](https://mlsanity.com/paper/2512.11526/) (2025)
- [The Wittgensteinian Representation Hypothesis: Is Language the Attractor of Multimodal Convergence?](https://mlsanity.com/paper/2605.09352/) (2026)
- [Learning Kernels by Alignment for Multiclass Bayes Classification](https://mlsanity.com/paper/2609.06474/) (2026)

## Abstract (arXiv)

It has recently been argued that AI models' representations are becoming aligned as their scale and performance increase. Empirical analyses have been designed to support this idea and conjecture the possible alignment of different representations toward a shared statistical model of reality. In this paper, we propose a learning-theoretic perspective to representation alignment. First, we review and connect different notions of alignment based on metric, probabilistic, and spectral ideas. Then, we focus on stitching, a particular approach to understanding the interplay between different representations in the context of a task. Our main contribution here is relating properties of stitching to the kernel alignment of the underlying representation. Our results can be seen as a first step toward casting representation alignment as a learning-theoretic problem.