---
task_categories:
- other
---

# When to Align, When to Predict: A Phase Diagram for Multimodal Learning

This repository contains pretrained encoders, cached features, and phase diagnostics data associated with the paper "[When to Align, When to Predict: A Phase Diagram for Multimodal Learning](https://huggingface.co/papers/2606.11190)".

The paper characterizes when contrastive (Cross-Alignment, CA) vs. predictive (Cross-Prediction, CP) self-supervised objectives recover shared signals in multimodal data, particularly for scientific domains.

[Project Website](https://ilaymalinyak.github.io/mm_align_vs_pred/) | [GitHub Repository](https://github.com/IlayMalinyak/mm_align_vs_pred)

## Dataset Description

The data stored here supports experiments including:
- **Astro Data**: Pretrained encoders and cached features for LAMOST × Kepler/TESS spectra.
- **Phase Diagnostics**: Data used to locate real-world datasets in the phase diagram regimes (Both, CA only, CP only, or Neither).

## Sample Usage

You can download the checkpoints, cached features, and phase diagnostics data using the `huggingface_hub` library:

```python
from huggingface_hub import snapshot_download
snapshot_download("Ilayk/mm_align_vs_pred", repo_type="dataset", local_dir="hf_data")
```

After downloading, you can point your environment to the data for the astrophysical experiments:
```bash
export MULTIDESA_ROOT=hf_data
```

## Citation

```bibtex
@article{kamai2026align,
  title   = {When to Align, When to Predict? A Phase Diagram for Multimodal Self-Supervised Learning},
  author  = {Kamai, Ilay and Van Assel, Hugues and Regev, Aviv and Perets, Hagai B. and Balestriero, Randall},
  journal = {arXiv preprint arXiv:2606.11190},
  year    = {2026}
}
```