Here it is people: test proving taumode 🍄🍄🍄
Since the publication of my latest paper I have received suggestions about testing my ideas on a real dataset. Here it is!
A complete unroll of the CVE dataset from 1999 to 2025 to:
⚙️ build a fine-tuned embedder on domain-specific text
🧩 generate a taumode index for the embeddings
❔❔❔ query the index
🧮 check the quality of the results against cosine similarity
The results demonstrate that we can search better than current vector databases do 👇👇👇
008_arrowspace_proof_of_concept_energy_informed_search – AI Research Engineering
Link to code available in the blog post.
Please consider sponsoring me on Github -> https://github.com/sponsors/Mec-iS
#vectorDB #embeddings #search #ranking #matching
taumode: a new way of searching vector databases was originally published in Coinmonks on Medium, where people are continuing the conversation by highlighting and responding to this story.