# Optimal Transport-based Alignment of Learned Character Representations for String Similarity

arXiv: https://arxiv.org/abs/1907.10165 · HTML: https://mlsanity.com/paper/1907.10165/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 23 Jul 2019.

## Built on by (most cited first, 7 of 7)

- [Multi-Vector Models with Textual Guidance for Fine-Grained Scientific Document Similarity](https://mlsanity.com/paper/2111.08366/) (2021)
- [Joint Optimal Transport and Embedding for Network Alignment](https://mlsanity.com/paper/2502.19334/) (2025)
- [Sublinear Time Approximation of Text Similarity Matrices](https://mlsanity.com/paper/2112.09631/) (2021)
- [Estimating the Performance of Entity Resolution Algorithms: Lessons Learned Through PatentsView.org](https://mlsanity.com/paper/2210.01230/) (2022)
- [A Riemannian Approach to Ground Metric Learning for Optimal Transport](https://mlsanity.com/paper/2409.10085/) (2024)
- [Neural String Edit Distance](https://mlsanity.com/paper/2104.08388/) (2021)
- [Optimal Transport for Handwritten Text Recognition in a Low-Resource Regime](https://mlsanity.com/paper/2509.16977/) (2025)

## Abstract (arXiv)

String similarity models are vital for record linkage, entity resolution, and search. In this work, we present STANCE --a learned model for computing the similarity of two strings. Our approach encodes the characters of each string, aligns the encodings using Sinkhorn Iteration (alignment is posed as an instance of optimal transport) and scores the alignment with a convolutional neural network. We evaluate STANCE's ability to detect whether two strings can refer to the same entity--a task we term alias detection. We construct five new alias detection datasets (and make them publicly available). We show that STANCE or one of its variants outperforms both state-of-the-art and classic, parameter-free similarity models on four of the five datasets. We also demonstrate STANCE's ability to improve downstream tasks by applying it to an instance of cross-document coreference and show that it leads to a 2.8 point improvement in B^3 F1 over the previous state-of-the-art approach.