Metraipolis › archive › dc_e226ae07509d

ADT: All-Atom Diffusion Tokenizer for 3D Molecular Structures — arXiv:2610.23265

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.23265
License (as archived)
CC0-1.0
Archived
2026-10-19T08:48:41.603Z
SHA-256 of the archived text
fc07c6a787c858aac8f2bfef4d84ea60425a217ffcc3590ab3fddeecb4fb19e0
Stable URL of this page
https://metraipolis.com/documento/dc_e226ae07509d

How to cite

ADT: All-Atom Diffusion Tokenizer for 3D Molecular Structures — arXiv:2610.23265 — arXiv (en), https://arxiv.org/abs/2610.23265 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e226ae07509d (Metraipolis archive dc_e226ae07509d, sha256 fc07c6a787c858aac8f2bfef4d84ea60425a217ffcc3590ab3fddeecb4fb19e0)
BibTeX
@misc{dce226ae07509d,
  title = {ADT: All-Atom Diffusion Tokenizer for 3D Molecular Structures — arXiv:2610.23265},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.23265},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e226ae07509d; sha256 fc07c6a787c858aac8f2bfef4d84ea60425a217ffcc3590ab3fddeecb4fb19e0},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23265. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Discrete structural tokenization provides a promising route to autoregressive 3D molecular generation. Yet, existing tokenizers face a trade-off between compact representation and all-atom reconstruction: atom-level tokens preserve atomic details but produce long sequences, while compact residue-level protein tokens typically omit side-chain geometry. To overcome this limitation, we propose All-atom Diffusion Tokenizer (ADT), which decouples token granularity from reconstruction resolution. ADT uses one structural token per atom for small molecules and one token per protein residue that jointly represents backbone and side-chain geometry. To mitigate geometric distortion introduced by quantization, we leverage the generative capacity of diffusion models to recover fine-grained atomic geometry beyond deterministic decoding from discrete tokens. Extensive experiments across small molecules and proteins demonstrate that ADT achieves strong reconstruction from compact discrete representations while enabling high-quality autoregressive generation of both small molecules and all-atom protein structures. These results establish compact structural tokens as an effective discrete interface for all-atom molecular reconstruction and generation.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.23265 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-19T08:42:40.269Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).