# Free-Grained Hierarchical Visual Recognition

Source: [Free-Grained Hierarchical Visual Recognition](<https://www.sota2.com/research/paper/68fa7c291e209131cea8cc5d>)

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

Data updated: 2026-05-22T05:00:12.726Z

Authors: Seulki Park, Zilin Wang, Stella X. Yu

Published: 2025-10-16T14:35:18.000Z

## Abstract

Hierarchical image recognition seeks to predict class labels along a semantic taxonomy, from broad categories to specific ones, typically under the tidy assumption that every training image is fully annotated along its taxonomy path. Reality is messier: A distant bird may be labeled only bird, while a clear close-up may justify bald eagle. We introduce free-grain training, where labels may appear at any level of the taxonomy and models must learn consistent hierarchical predictions from incomplete, mixed-granularity supervision. We build benchmark datasets with varying label granularity and show that existing hierarchical methods deteriorate sharply in this setting. To make up for missing supervision, we propose two simple solutions: One adds broad text-based supervision that captures visual attributes, and the other treats missing labels at specific taxonomy levels as a semi-supervised learning problem. We also study free-grained inference, where the model chooses how deep to predict, returning a reliable coarse label when a fine-grained one is uncertain. Together, our task, datasets, and methods move hierarchical recognition closer to the way labels arise in the real world.

## 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.

- [Hierarchical Visual Classification on CUB-F 13-38-200](<https://www.sota2.com/research/sota/hierarchical-visual-classification-on-cub-f-13-38-200.md>): Hierarchical Visual Classification; CUB-F 13-38-200; FPA: 63.96.

- [Hierarchical Visual Recognition on ImageNet-F 20-127-505 free-grain (test)](<https://www.sota2.com/research/sota/hierarchical-visual-recognition-on-imagenet-f-20-127-505-free-grain-test.md>): Hierarchical Visual Recognition; ImageNet-F 20-127-505 free-grain (test); FPA: 63.2.

- [Hierarchical Visual Recognition on iNat mini-F 21 free-grain (test)](<https://www.sota2.com/research/sota/hierarchical-visual-recognition-on-inat-mini-f-21-free-grain-test.md>): Hierarchical Visual Recognition; iNat mini-F 21 free-grain (test); FPA: 31.93.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2510.14737>)

[PDF](<https://arxiv.org/pdf/2510.14737.pdf>)