Honey, I shrunk the embeddings: Matryoshka vs. PCA

Discussion (6 Comments)Read Original on HackerNews

softwaredoug•29 minutes ago

Thanks for doing this benchmarking Dylan. I wanted to teach people PCA in my original article, but had no idea it would stack up this well against Matroyshka!

Feels like a “just use logistic regression” moment :)

stephantul•about 2 hours ago

Nice! I’ve been working on something similar and found similar results.

In my experiments, I used lots of embedding models and the results were not nearly as uniform as this curve, just FYI. I didn’t use any of the API-based models though

I also wrote about this exact comparison when using PCA and MRL to quantize static models, see: https://stephantul.github.io/blog/mrl-pca/

dcastm•42 minutes ago

Thank you! Will take a look at your results.

I couldn't find much when I first looked into this, which is why I ended up writing the article.

djoldman•14 minutes ago

Did you find much difference in inference latency or throughput between baseline and PCA?

stephantul•2 minutes ago

PCA is applied after the model, so there should be no difference in embedding throughput. Lookups in the index should be faster, but that speedup also applies equally to MRL.

So I guess the answer is: no

stephantul•40 minutes ago

Ah I meant more to say that I was working on this as well. I haven’t published the results for this comparison specifically yet.

Honey, I shrunk the embeddings: Matryoshka vs. PCA

⚡ Community Insights

Discussion (6 Comments)Read Original on HackerNews