# Extremely coarse learning objectives induce human-aligned representations in AI vision models

arXiv: https://arxiv.org/abs/2605.05556 · HTML: https://mlsanity.com/paper/2605.05556/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 70 references that are in the map.
- First posted on arXiv 7 May 2026.

## Builds on (most cited first, 24 of 24)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [DreamSim: Learning New Dimensions of Human Visual Similarity using Synthetic Data](https://mlsanity.com/paper/2306.09344/) (2023)
- [Brain-Like Object Recognition with High-Performing Shallow Recurrent ANNs](https://mlsanity.com/paper/1909.06161/) (2019)
- [Harmonizing the object recognition strategies of deep neural networks with humans](https://mlsanity.com/paper/2211.04533/) (2022)
- [Deep Neural Networks Rival the Representation of Primate IT Cortex for Core Visual Object Recognition](https://mlsanity.com/paper/1406.3284/) (2014)
- [The neuroconnectionist research programme](https://mlsanity.com/paper/2209.03718/) (2022)
- [Improving neural network representations using human similarity judgments](https://mlsanity.com/paper/2306.04507/) (2023)
- [Performance-optimized deep neural networks are evolving into worse models of inferotemporal visual cortex](https://mlsanity.com/paper/2306.03779/) (2023)
- [On the Trade-off of Intra-/Inter-class Diversity for Supervised Pre-training](https://mlsanity.com/paper/2305.12224/) (2023)
- [Understanding the Impact of Label Granularity on CNN-based Image Classification](https://mlsanity.com/paper/1901.07012/) (2019)
- [Statistical inference on representational geometries](https://mlsanity.com/paper/2112.09200/) (2021)
- [Scaling Laws for Task-Optimized Models of the Primate Visual Ventral Stream](https://mlsanity.com/paper/2411.05712/) (2024)
- [Universal dimensions of visual representation](https://mlsanity.com/paper/2408.12804/) (2024)
- [Weakly Supervised Representation Learning with Coarse Labels](https://mlsanity.com/paper/2005.09681/) (2020)
- [Semi-Supervised Learning with Taxonomic Labels](https://mlsanity.com/paper/2111.11595/) (2021)
- [Coarse-to-Fine Curriculum Learning](https://mlsanity.com/paper/2106.04072/) (2021)
- [Basic Level Categorization Facilitates Visual Object Recognition](https://mlsanity.com/paper/1511.04103/) (2015)
- [Learning Hierarchical Visual Representations in Deep Neural Networks Using Hierarchical Linguistic Labels](https://mlsanity.com/paper/1805.07647/) (2018)
- [Vision CNNs trained to estimate spatial latents learned similar ventral-stream-aligned representations](https://mlsanity.com/paper/2412.09115/) (2024)
- [The Advantage of Fine-Grained Training](https://mlsanity.com/paper/2509.05130/) (2025)

## Built on by (most cited first, 1 of 1)

- [Characterizing the visual representation of objects from the child's view](https://mlsanity.com/paper/2605.14990/) (2026)

## Abstract (arXiv)

Artificial neural networks trained on visual tasks develop internal representations resembling those of the primate visual system, a discovery that has guided a decade of computational neuroscience. Research on building brain-aligned models has progressively embraced finer-grained learning ob- jectives, from object classification to contrastive self-supervised objectives that maximize distinc- tions among individual images. Yet the effect of learning-signal granularity on brain alignment remains largely unexamined. Here we systematically investigate how the granularity of a learning signal shapes representational alignment with human vision. We parametrically vary the number of training classes using a data-driven approach that partitions a set of training images into differ- ent numbers of categories via PCA-based splits of pretrained embeddings. We train hundreds of neural networks across convolutional and transformer architectures on these coarse classification tasks and compare their representations with human fMRI responses, macaque electrophysiology recordings, and human behavior. We find that networks trained to distinguish as few as eight broad categories learn representations that match or exceed the neural alignment of models distinguishing 1,000 classes. Even more strikingly, these coarsely trained networks align more closely with hu- man perceptual similarity judgments than all other models evaluated, including networks trained with fine-grained supervision or self-supervision as well as leading large-scale vision models. These results demonstrate that human-like visual representations can emerge from surprisingly simple learning objectives, reframing what learning signals vision may require and opening a path toward building AI systems that are more aligned with human perception.