| title | Balancing Fidelity and Diversity in Diffusion Models via Symmetric Attention Decomposition: Hopfield Perspective |
|---|---|
| authors | Hyunmin Cho, Woo Kyoung Han, Kyong Hwan Jin |
| date | 2026-05-26 |
| venue | arXiv (preprint, not peer reviewed) |
| id | arXiv:2605.27476 |
| upvotes | 13 (Hugging Face Daily Papers, 2026-11-01) |
| links | abs pdf hf code |
| abstract | We characterize the pre-softmax attention matrix QK^top in transformers as an associative memory matrix encoding pairwise associations between input features. By decomposing this matrix into its symmetric and skew-symmetric parts, we interpret the symmetric component as governing the structure of the energy landscape, and the skew-symmetric component as driving circulation on that landscape. Leveraging the energy formulation induced by the symmetric component, we derive Hopfield-style stability measures that quantify the stability of retrieved features. We observe meaningful correlations between Hopfield-style stability measures and the fidelity-diversity trade-offs in generation. Finally, we propose a controllable knob to modulate this trade-off by modifying the circulation of the underlying dynamics. Code is available at our GitHub (https://github.com/hyeon-cho/Attention-Symmetric-Decomposition). |
cite
@misc{cho2026balancing,
title={Balancing Fidelity and Diversity in Diffusion Models via Symmetric Attention Decomposition: Hopfield Perspective},
author={Hyunmin Cho and Woo Kyoung Han and Kyong Hwan Jin},
year={2026},
eprint={2605.27476},
archivePrefix={arXiv},
url={https://arxiv.org/abs/2605.27476}
}