Looking for Model Recommendations to Embed 10-K and 10-Q SEC Filings for RAG System

#144
by sarra10 - opened

i'm currently working on building a Retrieval-Augmented Generation (RAG) system focused on analyzing 10-K and 10-Q SEC filings. My goal is to find a robust embedding model that can handle the dense, domain-specific language in these financial documents and generate meaningful embeddings for efficient retrieval.If anyone here has experience with embeddings for similar use cases or knows of models that excel in processing financial documents, I'd appreciate your insights

Massive Text Embedding Benchmark org

Hi @sarra10 we don't take issues on this space (see pinned issue).

But you are free to ask usage questions here: https://github.com/embeddings-benchmark/mteb/discussions

Also, note that we have a PR for a finance benchmark coming up, but it is currently a work in progress.

KennethEnevoldsen changed discussion status to closed

Sign up or log in to comment