mlsanityData to 7 Oct 2026

mini-vec2vec: Scaling Universal Geometry Alignment with Linear Transformations

arXiv 2510.02348 ↗First posted 27 Sep 2025cs.CL cs.AI cs.LGcited by 3

Not on the map. Why?

Abstract (arXiv)

We build upon vec2vec, a procedure designed to align text embedding spaces without parallel data. vec2vec finds a near-perfect alignment, but it is expensive and unstable. We present mini-vec2vec, a simple and efficient alternative that requires substantially lower computational cost and is highly robust. Moreover, the learned mapping is a linear transformation. Our method consists of three main stages: a tentative matching of pseudo-parallel embedding vectors, transformation fitting, and iterative refinement. Our linear alternative exceeds the original instantiation of vec2vec by orders of magnitude in efficiency, while matching or exceeding their results. The method's stability and interpretable algorithmic steps facilitate scaling and unlock new opportunities for adoption in new domains and fields.

Key facts. Cited by 3 papers in the map (data through 7 Oct 2026). Builds on 25 of its 39 references that are in the map. First posted on arXiv 27 Sep 2025.

Builds on

25 of 39 references.

References · 25

  1. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,216
  2. 2023Towards General Text Embeddings with Multi-stage Contrastive Learningcited by 705
  3. 2017Word Translation Without Parallel Datacited by 480
  4. 2021Large Dual Encoders Are Generalizable Retrieverscited by 454
  5. 2024The Platonic Representation Hypothesiscited by 362
  6. 2016NIPS 2016 Tutorial: Generative Adversarial Networkscited by 284
  7. 2017Unsupervised Machine Translation Using Monolingual Corpora Onlycited by 272
  8. 2014Understanding image representations by measuring their equivariance and equivalencecited by 240
  9. 2021Revisiting Model Stitching to Compare Neural Representationscited by 203
  10. 2020Information Leakage in Embedding Modelscited by 199
  11. 2023Text Embeddings Reveal (Almost) As Much As Textcited by 189
  12. 2013Exploiting Similarities among Languages for Machine Translationcited by 187
13 more
  1. 2022Relative representations enable zero-shot latent space communicationcited by 175
  2. 2017Unsupervised Neural Machine Translationcited by 148
  3. 2018A robust self-learning method for fully unsupervised cross-lingual mappings of word embeddingscited by 145
  4. 2022Analyzing Transformers in Embedding Spacecited by 131
  5. 2018Gromov-Wasserstein Alignment of Word Embedding Spacescited by 119
  6. 2024Jasper and Stella: distillation of SOTA embedding modelscited by 109
  7. 2018Unsupervised Alignment of Embeddings with Wasserstein Procrustescited by 96
  8. 2025Harnessing the Universal Geometry of Embeddingscited by 54
  9. 2024Do Vision and Language Encoders Represent the World Similarly?cited by 44
  10. 2018Non-Adversarial Unsupervised Word Translationcited by 19
  11. 2023Discovering Universal Geometry in Embeddings with ICAcited by 19
  12. 2018Unsupervised Multilingual Word Embeddingscited by 16
  13. 2025It's a (Blind) Match! Towards Vision-Language Correspondence without Parallel Datacited by 13

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 3.

3 papers.

  1. 2026Back into Plato's Cave: Examining Cross-modal Representational Convergence at Scalecited by 10
  2. 2026Platonic Representations in the Human Brain: Unsupervised Recovery of Universal Geometrycited by 0
  3. 2026Unpaired Canonical Correlation Analysiscited by 0