# Atelier: Learning Local Self-Supervised Features for CryoEM Volumes via Hypernetworks

arXiv: https://arxiv.org/abs/2609.30569 · HTML: https://mlsanity.com/paper/2609.30569/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 64 references that are in the map.
- First posted on arXiv 24 Sep 2026.

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

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [A Simple Framework for Contrastive Learning of Visual Representations](https://mlsanity.com/paper/2002.05709/) (2020)
- [Masked Autoencoders Are Scalable Vision Learners](https://mlsanity.com/paper/2111.06377/) (2021)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [FlashAttention: Fast and Memory-Efficient Exact Attention with IO-Awareness](https://mlsanity.com/paper/2205.14135/) (2022)
- [Implicit Neural Representations with Periodic Activation Functions](https://mlsanity.com/paper/2006.09661/) (2020)
- [Fourier Features Let Networks Learn High Frequency Functions in Low Dimensional Domains](https://mlsanity.com/paper/2006.10739/) (2020)
- [HyperNetworks](https://mlsanity.com/paper/1609.09106/) (2016)
- [Scene Representation Networks: Continuous 3D-Structure-Aware Neural Scene Representations](https://mlsanity.com/paper/1906.01618/) (2019)
- [UNet++: Redesigning Skip Connections to Exploit Multiscale Features in Image Segmentation](https://mlsanity.com/paper/1912.05074/) (2019)
- [Learning Continuous Image Representation with Local Implicit Image Function](https://mlsanity.com/paper/2012.09161/) (2020)
- [DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition](https://mlsanity.com/paper/1310.1531/) (2013)
- [Decomposing NeRF for Editing via Feature Field Distillation](https://mlsanity.com/paper/2205.15585/) (2022)
- [MetaSDF: Meta-learning Signed Distance Functions](https://mlsanity.com/paper/2006.09662/) (2020)
- [From data to functa: Your data point is a function and you can treat it like one](https://mlsanity.com/paper/2201.12204/) (2022)
- [Adversarial Generation of Continuous Images](https://mlsanity.com/paper/2011.12026/) (2020)
- [Transformers as Meta-Learners for Implicit Neural Representations](https://mlsanity.com/paper/2208.02801/) (2022)
- [CryoBench: Diverse and challenging datasets for the heterogeneity problem in cryo-EM](https://mlsanity.com/paper/2408.05526/) (2024)
- [Magnitude Invariant Parametrizations Improve Hypernetwork Learning](https://mlsanity.com/paper/2304.07645/) (2023)
- [Hypernetworks build Implicit Neural Representations of Sounds](https://mlsanity.com/paper/2302.04959/) (2023)
- [Pearson-Matthews correlation coefficients for binary and multinary classification and hypothesis testing](https://mlsanity.com/paper/2305.05974/) (2023)
- [A Graph Neural Network Approach to Automated Model Building in Cryo-EM Maps](https://mlsanity.com/paper/2210.00006/) (2022)
- [Acquiring and Adapting Priors for Novel Tasks via Neural Meta-Architectures](https://mlsanity.com/paper/2507.10446/) (2025)
- [CryoHype: Reconstructing a thousand cryo-EM structures with transformer-based hypernetworks](https://mlsanity.com/paper/2512.06332/) (2025)

## Abstract (arXiv)

CryoEM map interpretation requires features that are spatially localized, consistent across samples, and informative across spatial scales. Most deep learning methods for map annotation extract features from fixed voxel grids. However, implicit neural representations (INRs) are able to model volumetric data as scale-agnostic, coordinate-conditioned functions. INRs are therefore attractive for cryoEM, but fitting a separate INR for each map is too expensive for large-scale feature extraction and produces representations that are not aligned across samples. We introduce Atelier, a self-supervised framework that amortizes INR fitting for reconstructed cryoEM maps. Pretrained on 5,439 Electron Microscopy Data Bank maps, Atelier is a transformer-based hypernetwork that generates high-fidelity reconstructions across a wide range of protein structures, including large multi-subunit assemblies. Beyond reconstruction, the INR generated by the pretrained transformer exposes a continuous, local feature field through its intermediate activations at any spatial query point, a property that voxel grid and patch-tokenizer architectures do not naturally provide. Used as auxiliary channels to a 3D nested U-Net annotation head trained from scratch, these coordinate-conditioned features improve performance on eight voxel-level property prediction tasks over a volume-only baseline. Our results demonstrate that amortized implicit neural representations are an effective primitive for geometry-aware analysis of cryoEM data.