# Combining Different Latent Spaces?

**URL:** <https://community.openai.com/t/combining-different-latent-spaces/1357813>\
**Category:** Open Models\
**Tags:** embeddings\
**Created:** [September 12, 2025, 5:15pm UTC](https://community.openai.com/t/combining-different-latent-spaces/1357813 "2025-09-12T17:15:15Z")\
**Posts on this page:** 2\
**Page:** 1

<div class="post-metadata">

**Author:** ![jack.kausch](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/jack.kausch/32/678885_2.png) [@jack.kausch](https://community.openai.com/u/jack.kausch)\
**Post date:** [September 12, 2025, 5:15pm UTC](https://community.openai.com/t/combining-different-latent-spaces/1357813/1 "2025-09-12T17:15:15Z")

</div>

Hi,

I found a post here about comparing cosine similarities over time.

I am doing something similar but a bit different. I have created an information retrieval engine that operates over document embeddings in different languages, reusing the same categories to index those embeddings in a projected feature space of cosine similarities.

There’s a faceted browser which allows users to combine features in a context sensitive way and browse the pooled cosine similarities.

My question is: does the community think this could have any use for aligning document embeddings in different languages?

i.e., if I reused the same features and indexed a large number of vector databases with anchor vectors that represented those features, and pooled them continuously until I had one very large cosine-similarity feature space, would it be useful for combining documents, or would it just contain meaningless noise and strange artefacts? Could I use it to align different embedding spaces over time? Would it be useful for training a model or for RAG?

Or is this approach just extremely naive?

---

<div class="post-metadata">

**Author:** ![Macha](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/macha/32/171757_2.png) [@Macha](https://community.openai.com/u/Macha)\
**Post date:** [September 20, 2025, 10:49pm UTC](https://community.openai.com/t/combining-different-latent-spaces/1357813/2 "2025-09-20T22:49:14Z")

</div>

Hey there and welcome!

Sounds like a neat project.

> [@jack.kausch](#):
>
> I have created an information retrieval engine that operates over document embeddings in different languages,

Just to be clear here, do you mean like, different _natural_ languages that represent the contents of the docs, or different _programming_ languages here?

> [@jack.kausch](#):
>
> i.e., if I reused the same features and indexed a large number of vector databases with anchor vectors that represented those features, and pooled them continuously until I had one very large cosine-similarity feature space, would it be useful for combining documents, or would it just contain meaningless noise and strange artefacts?

Perhaps yes, but perhaps this might also be rather over-complicating something that could be solved with a simple graph database? You can essentially just make your own pool of vectorized data that has any kind of arbitrary connections/clusters or links between nodes that you’d like. Myself and others here really like [neo4j](https://neo4j.com/) because of the cool ways in which you can cluster and vectorize things for easy RAG with an LLM.