Metraipolis › archive › dc_4ae4fbf86c74

Let CSP Be Your ANCHOR: Adaptive Crystal Search over Frozen Structure Priors — arXiv:2609.33407

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.33407
License (as archived)
CC0-1.0
Archived
2026-10-07T22:40:00.397Z
SHA-256 of the archived text
3c623adbdfaee65195ae78e950323c34c51ff1acf5cbd5ca6d8bc746856eb2d4
Stable URL of this page
https://metraipolis.com/documento/dc_4ae4fbf86c74

How to cite

Let CSP Be Your ANCHOR: Adaptive Crystal Search over Frozen Structure Priors — arXiv:2609.33407 — arXiv (en), https://arxiv.org/abs/2609.33407 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_4ae4fbf86c74 (Metraipolis archive dc_4ae4fbf86c74, sha256 3c623adbdfaee65195ae78e950323c34c51ff1acf5cbd5ca6d8bc746856eb2d4)
BibTeX
@misc{dc4ae4fbf86c74,
  title = {Let CSP Be Your ANCHOR: Adaptive Crystal Search over Frozen Structure Priors — arXiv:2609.33407},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.33407},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_4ae4fbf86c74; sha256 3c623adbdfaee65195ae78e950323c34c51ff1acf5cbd5ca6d8bc746856eb2d4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.33407. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

De novo crystal generation (DNG) models decide where to search in composition space and how to generate structures with one set of weights. We argue that discovery is better served by separating the two. A crystal structure prediction (CSP) model is a physical prior that should be improved by likelihood training, while rewards, including novelty measured against the search's own history, should act on a search over compositions. We introduce ANCHOR, a GRPO composition policy trained with multi-objective rewards around a frozen CSP model, and continuous adaptive novelty (CAN), a graded novelty score against known structures and a growing discovery history. Using the frozen CSP model as a fixed ruler under one evaluator, we test where adaptation should act. Replacing DNG compositions with ANCHOR's policy on the same CSP backbone raises MSUN from 11.4% to 47.6% and SUN from 1.1% to 22.1% at 99.9% formula uniqueness. Fine-tuning DNG models directly on the same rewards instead moves their composition marginal without raising their on-hull fraction. We show that KL-regularized fine-tuning of a DNG model can only reweight chemistry the pretrained model already supports by a bounded factor, while unregularized DNG fine-tunes move toward known or less stable chemistry. Even a stability-only reward routed into ANCHOR's CSP backbone roughly halves SUN relative to the frozen backbone, whereas likelihood training on structures found during search can improve a CSP backbone. Under MatterGen's evaluation pipeline, ANCHOR raises state-of-the-art MSUN from 29.2% to 41.3%, transfers without retraining to two further CSP backbones, and reaches 47.1% after distillation into Crystalite-CSP. As with any model optimised against a potential, its on-hull rate depends on that potential.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-29 in cs.LG).
Categorie arXiv: Materials Science (cond-mat.mtrl-sci) ; Machine Learning (cs.LG); Computational Physics (physics.comp-ph).
Pagina del lavoro: https://arxiv.org/abs/2609.33407 — 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 2026-10-07T22:14:28.715Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-29?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).