arXiv cs.AI· Tian Luo, Ruge Zhang, Haozhi Han, Yifeng Chen, Yunquan Zhang, Yunxin Liu, Ting Cao, Kun Li·· 12 小时前AI 评分31
AtomWorld-Mem：面向长时程原子演化的记忆恢复世界状态模型
AtomWorld-Mem: Memory-Restored World States for Long-Horizon Atomistic Evolution
AI 导读
AtomWorld-Mem 将长时程原子演化建模为基于记忆的世界状态恢复问题，通过空间编码器写入多尺度原子关键帧，并结合短期事件记忆与长期结构记忆恢复瞬时快照中缺失的隐藏状态。
正文 · 原文
Abstract:High-fidelity atomistic evolution over long timescales requires more than observing the current crystal configuration. Instantaneous atomistic snapshots are often incomplete: locally similar configurations can correspond to different hidden dynamical contexts, future event preferences, and waiting-time scales. We argue that this snapshot ambiguity makes long-horizon atomistic evolution fundamentally a memory-based world-state restoration problem. To address this, we introduce AtomWorld-Mem, a memory-restored atomistic world model that recovers the latent world state missing from instantaneous crystal snapshots. AtomWorld-Mem treats the evolving alloy as an AtomWorld: spatial encoders write multi-scale atomistic keyframes from dense local topology and sparse long-range defect context, while short-term event memory and long-term structural memory integrate these keyframes across time to restore a future-predictive evolutionary state. The restored state is used to prioritize legal vacancy-mediated events under single-event Kinetic Monte Carlo (KMC) constraints, while event legality, physical execution, and residence-time updates remain governed by the underlying simulator. Empirically, AtomWorld-Mem improves long-horizon atomistic progress under fixed microscopic event budgets while maintaining high-fidelity evolution across energetic, structural, and vacancy-transport observables. It further transfers zero-shot across diverse unseen alloy-temperature AtomWorlds, suggesting that the learned memory-restoration mechanism captures reusable principles of hidden-state inference rather than a system-specific local energy heuristic. These results position memory-restored world-state modeling as a promising route toward efficient, physically grounded, and transferable atomistic evolution.
| Subjects: | Artificial Intelligence (cs.AI); Materials Science (cond-mat.mtrl-sci) |
| Cite as: | arXiv:2609.31133 [cs.AI] |
| (or arXiv:2609.31133v3 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2609.31133 arXiv-issued DOI via DataCite |
Submission history
From: Tian Luo [view email]
[v1]
Fri, 25 Sep 2026 11:26:05 UTC (3,192 KB)
[v2]
Sun, 4 Oct 2026 06:02:12 UTC (3,192 KB)
[v3]
Tue, 20 Oct 2026 11:08:38 UTC (3,203 KB)
来源：arXiv cs.AI · arxiv.org