# Vision encoders should be image size agnostic and task driven

Source: [Vision encoders should be image size agnostic and task driven](<https://www.sota2.com/research/paper/68fa7aca1e209131cea8519b>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-26T05:26:48.092Z

Authors: Nedyalko Prisadnikov, Danda Pani Paudel, Yuqian Fu, Luc Van Gool

Published: 2025-08-22T11:57:49.000Z

## Abstract

This position paper argues that the next generation of vision encoders should be image size agnostic and task driven. The source of our inspiration is biological. Not a structural aspect of biological vision, but a behavioral trait -- efficiency. We focus on a couple of ways in which vision in nature is efficient, but modern vision encoders not. We -- humans and animals -- deal with vast quantities of visual data, and need to be smart where we focus our limited energy -- it depends on the task. It is our belief that vision encoders should be dynamic and the computational complexity should depend on the task at hand rather than the size of the image. We, also, provide concrete first steps towards our vision -- a proof-of-concept solution for image classification. Despite classification being not very representative for what we are trying to achieve, it shows that our approach is feasible and promising.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2508.16317>)

[PDF](<https://arxiv.org/pdf/2508.16317.pdf>)