Xu Qi | Searcharxiv

arXiv2026

Hierarchical Relaxation in Large-twist-angle Moire Drives Supermoire Formation

Supermoire lattices in large-angle twisted bilayers have recently been shown to emerge from periodic arrangements of commensurate stacking motifs, yet the microscopic structural mechanism governing their formation has remained unresolved. Here, we uncover the microscopic origin of supermoire order as a hierarchical lattice relaxation process. Combining force-field simulations with experimental validation via super-resolution, layer-resolved multislice electron ptychography, our results identify interlayer Se-Se repulsion at the picometer scale as the driving force, giving rise to coupled short-range (about 1 nm) moire relaxation and coexisting long-range (about 10 nm) supermoire modulation. These findings demonstrate that even picometer-scale lattice relaxation remains critical in large-angle moire systems and must be explicitly considered in understanding their emergent electronic and quantum phenomena.

cond-mat.mtrl-sci↗

arXiv2024

UnifiedMLLM: Enabling Unified Representation for Multi-modal Multi-tasks With Large Language Model

Significant advancements has recently been achieved in the field of multi-modal large language models (MLLMs), demonstrating their remarkable capabilities in understanding and reasoning across diverse tasks. However, these models are often trained for specific tasks and rely on task-specific input-output formats, limiting their applicability to a broader range of tasks. This raises a fundamental question: Can we develop a unified approach to represent and handle different multi-modal tasks to maximize the generalizability of MLLMs? In this paper, we propose UnifiedMLLM, a comprehensive model designed to represent various tasks using a unified representation. Our model exhibits strong capabilities in comprehending the implicit intent of user instructions and preforming reasoning. In addition to generating textual responses, our model also outputs task tokens and grounding tokens, serving as indicators of task types and task granularity. These outputs are subsequently routed through the task router and directed to specific expert models for task completion. To train our model, we construct a task-specific dataset and an 100k multi-task dataset encompassing complex scenarios. Employing a three-stage training strategy, we equip our model with robust reasoning and task processing capabilities while preserving its generalization capacity and knowledge reservoir. Extensive experiments showcase the impressive performance of our unified representation approach across various tasks, surpassing existing methodologies. Furthermore, our approach exhibits exceptional scalability and generality. Our code, model, and dataset will be available at \url{https://github.com/lzw-lzw/UnifiedMLLM}.

cs.CL↗