File size: 1,984 Bytes
28d72ba c837520 28d72ba c837520 28d72ba c837520 28d72ba c837520 28d72ba c837520 | 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 | ---
library_name: tedbench
license: bsd-3-clause
pipeline_tag: graph-ml
tags:
- protein
- structure
- fold-classification
- tedbench
---
# TEDBench — Supervised from scratch (structure + sequence)
**Variant:** `miae_b` + seq | **Parameters:** 102M | **Layers:** 12 | **Hidden dim:** 768 | **Attn heads:** 12
This **MiAEClassifier** was trained from scratch on TEDBench without pretraining. This model is part of the work presented in the paper [Protein Fold Classification at Scale: Benchmarking and Pretraining](https://arxiv.org/abs/2605.18552).
> **+seq variant** — sequence embeddings are concatenated to the geometric encoder input (`model.use_seq_input=true`).
Part of the [TEDBench](https://github.com/BorgwardtLab/TEDBench) benchmark for protein fold classification (ICML 2026). MiAE is an SE(3)-invariant masked autoencoder that masks up to 90% of backbone frames and reconstructs the full structure with a lightweight decoder.
## Architecture sizes
| Variant | Params | Layers | Hidden dim | Attn heads |
|---------|-------:|-------:|-----------:|-----------:|
| `miae_s` | 29 M | 6 | 512 | 8 |
| `miae_b` | 102 M | 12 | 768 | 12 |
| `miae_l` | 339 M | 24 | 1 024 | 16 |
Append `+model.use_seq_input=true` to `miae_b` for the **+seq** variant.
## Usage
### Load from the HuggingFace Hub
```python
from tedbench.utils.io import load_from_hf
model = load_from_hf("TEDBench/miae-b-seq-sc")
model.eval()
```
### From a Lightning checkpoint
```python
from tedbench.model import MiAEClassifier
model = MiAEClassifier.load_from_checkpoint("model.ckpt", weights_only=False)
model.eval()
```
## Citation
```bibtex
@inproceedings{chen2026tedbench,
title={Protein Fold Classification at Scale: Benchmarking and Pretraining},
author={Chen, Dexiong state, Andrei Manolache, Mathias Niepert, Karsten Borgwardt},
booktitle={Proceedings of the 43rd International Conference on Machine Learning},
year={2026}
}
``` |