Feature Request: support for XLMRobertaModel embeddings

[IzzyHibbert]

Prerequisites

• I am running the latest code. Mention the version if possible as well.
• I carefully followed the README.md.
• I searched using keywords relevant to my issue to make sure that I am creating a new issue that is not already open (or closed).
• I reviewed the Discussions, and have a new and useful enhancement to share.

Feature Description

At the moment the project does not support embeddings based on XLMRobertaModel architecture.
Could be very good to extend the support to those.

Motivation

I believe that adding support could be great. Models like BAAI/bge-m3 are showing promising results against proprietary models like OpenAI, see here :
https://towardsdatascience.com/openai-vs-open-source-multilingual-embedding-models-e5ccb7c90f05

Also, they support higher volume of tokens (bge--m3 for instance up to 8192 tokens).

Possible Implementation

No response

[ExtReMLapin]

#8658

[IzzyHibbert]

Closed as duplicate of #8658. (apologies, did not find it when searching..)