Latent space configuration for neural network finetuning, fast ranking, and embedding dimensionality reduction

Nikita Gabdullin, Anton Raskovalov
10/21/2026
cs.CVcs.AI

No confident code implementation found for this paper.

Abstract

Finding K most similar elements for a query in a large gallery of neural network (NN) embeddings is essential in various areas of computer science and industrial applications of artificial intelligence. This paper proposes search algorithms for structured latent spaces (LS) of NNs which allow to speed up top K item retrieval owing to O(1) complexity small candidate gallery subset search. ArcFace-pretrained IResNet50 models are finetuned using Latent Space Configuration (LSC) methodology to obtain new distribution of embeddings with desired geometric properties. Applying the proposed algorithms to embeddings obtained from finetuned models leads to several times speed-up over baseline exhaustive search. Specifically, we report up to 3 times speed-up on GPU and 40% speed-up on CPU compared to the baseline method GPU speed. These results are obtained for the original and reduced numbers of LS dimensions. The results show that LS dimensionality reduction allows to further accelerate the search and even improve rank-N metrics.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

sharpDARTS: Faster and More Accurate Differentiable Architecture Search

85% match1 repo152019

Automated Search-Space Generation Neural Architecture Search

82% match1 repo3112023

Latent space configuration for improved generalization in supervised autoencoder neural networks

81% match2024

Sensitivity-Aware Mixed-Precision Quantization and Width Optimization of Deep Neural Networks Through Cluster-Based Tree-Structured Parzen Estimation

81% match2023

Series of quasi-uniform scatterings with fast search, root systems and neural network classifications

81% match2025

FMAS: Fast Multi-Objective SuperNet Architecture Search for Semantic Segmentation

80% match2023

Cite this paper

@article{gabdullin2026latent,
  title  = {Latent space configuration for neural network finetuning, fast ranking, and embedding dimensionality reduction},
  author = {Nikita Gabdullin and Anton Raskovalov},
  year   = {2026},
  eprint = {2610.29295},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2610.29295v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2610.29295.svg)](https://codeofpaper.com/paper/2610.29295)

Discussion