Vikasht34
Vikasht34
@benwtrent haha , It looks like from LLM , Originally it came from this paper https://www.arxiv.org/pdf/1908.00814v5 ``` The upper layer graph is merged into the approximate k-NN graph of the...
@benwtrent Should we create separate issue and I can take a stab , implementing it with this paper https://www.arxiv.org/pdf/1908.00814v5
@benwtrent Thanks ...
I am owner for 2.18 release , somehow this ticket is not getting assigned to me.
Hi @rchitale7 Thanks for putting this up, I have 2 major concern on whole design. 1. Default In memory Request Store (HashMap based Implementation) :- As a general concept from...
> he worker is stateless. A stateful system doesn't mean that it has to store the information in the memory. > > A simple difference between stateful and stateless system:...
> @Vikasht34 For this, I think we probably can break up the random rotation and the computation of mean centers into 2 separate features. I think that they can be...
This seems to be a very interesting discussion, The only realistic safe approach is conditional restore like: Allow backward restore only if: - The snapshot’s Lucene version == target’s Lucene...