arXiv subjects

Zi-Ang Wang

Publications and source records attributed to Zi-Ang Wang.

5 recordsLinked to original sources

arXiv2026

Energy Gap in Weakly Disordered Fractional Quantum Hall Liquids: Quantitative Comparison to GaAs Quantum Well Experiments at $\nu= 1/3$

Based on a recent experiment in high-quality GaAs quantum wells [Phys. Rev. Lett. 127, 056801 (2021)], we present a microscopic study of the energy gap in two-dimensional electron gases at filling factor $\nu=1/3$, explicitly incorporating both finite layer thickness and disorder effects. The finite layer thickness is modeled by solving the Poisson-Schr\"odinger equations for the experimental devices, yielding the electron wave functions in the perpendicular direction. Using these and the disorder energy extracted from the experiment, we estimate the charge gap and the mobility gap at $\nu=1/3$ in the weakly disordered lowest Landau level. Remarkably, both gaps show good quantitative agreement with the activation gap measured from the experiment in narrow quantum wells. Our results also indicate the potential need of incorporating higher subbands to make accurate theoretical predictions of the energy gap in wide quantum wells.

cond-mat.str-el↗

arXiv2025

Tuning Bound States of Symmetry-Breaking Vortices via Unidirectional Charge Density Wave in a Transition-Metal Dichalcogenide Superconductor

The interplay between charge density wave (CDW) and superconducting vortex bound states are crucial for fundamental physics of superconductivity and advancing quantum nanotechnologies. However, the CDW-mediated modulation of vortex bound states, which opens up a new platform for vortex engineering, remains unexplored. Here, we report spatially anisotropic vortex states modulated by the unidirectional CDWs in a transition-metal dichalcogenide superconductor 1T''-NbTe2 using ultra-low-temperature scanning tunneling microscopy/spectroscopy. The stripe-like 3x1x3 CDW order exhibits a robust three-dimensional character across step edges and coexists with superconductivity below a critical temperature of 0.4 K. Under out-of-plane magnetic fields, we observe elliptical vortices whose elongation aligns with the CDW stripes, indicating strong coupling between vortex morphology and underlying electronic order. Remarkably, CDW domain boundaries induce abrupt changes in vortex orientation and vortex bound states, enabling controllable vortex states across CDW nanodomains. These findings establish a new pathway for manipulating superconducting vortex bound states via CDW coupling.

cond-mat.supr-con↗

arXiv2025

Atomic Visualization of Bulk and Surface Superconductivity in Weyl Semimetal {\gamma}-PtBi2

A bulk superconductor hosting intrinsic surface superconductivity provides a unique platform to study Majorana bound states. The superconductor, trigonal {\gamma}-PtBi2, is a promising candidate, as surface superconducting gaps and topological surface states have been observed. However, the simultaneous presence of bulk and surface superconductivity has not been resolved. Here, we directly visualize coexisting bulk and surface superconducting gaps in trigonal PtBi2 by using ultra-low-temperature scanning tunneling microscopy/spectroscopy. The bulk gap is {\Delta} ~ 0.053 meV with a critical temperature (Tc) ~ 0.5 K and a critical field below 0.01 T, accompanied by a vortex lattice and bound states, yielding a coherence length of ~200 nm. Remarkably, certain surface regions show a much larger gap of {\Delta} ~ 0.42 meV, persisting up to Tc ~ 3 K and surviving magnetic fields up to 2 T, yet lacking a static vortex lattice. This coexistence of robust surface and bulk superconductivity establishes {\gamma}-PtBi2 as a unique platform for investigating the interplay between bulk and surface Cooper pairings in superconducting topological materials.

cond-mat.supr-con↗

arXiv2025

Recent progress of scanning tunneling microscopy/spectroscopy study of pair density wave in superconductors

A pair density wave (PDW) is a superconducting state characterized by an order parameter with finite center-of-mass momentum in the absence of an external magnetic field, thereby breaking the conventional translational symmetry in homogeneous superconductors. It is proposed that PDW emerges from magnetic interactions, strong electron-electron correlations, and their interplay with competing orders. In this review, we highlight recent advances in the detection and study of PDWs using scanning tunneling microscopy and spectroscopy (STM/STS). We focus on how the signatures of PDW have been experimentally visualized across a variety of extraordinary superconductors, including iron-based superconductors, cuprate superconductors, spin-triplet superconductors, kagome-lattice superconductors, and transition metal dichalcogenides. Beginning with an introduction to the fundamental concept of PDWs and the unique capabilities of STM/STS, particularly its atomic-scale spatial resolution and advanced data analysis techniques, we discuss key experimental findings, including the direct visualization of charge density modulations associated with PDWs. Finally, we discuss emerging challenges and future directions, aiming to inspire future research into the nature of PDWs in superconductors.

cond-mat.supr-con↗

arXiv2025

Synthetic Data RL: Task Definition Is All You Need

Reinforcement learning (RL) is a powerful way to adapt foundation models to specialized tasks, but its reliance on large-scale human-labeled data limits broad adoption. We introduce Synthetic Data RL, a simple and general framework that reinforcement fine-tunes models using only synthetic data generated from a task definition. Our method first generates question and answer pairs from the task definition and retrieved documents, then adapts the difficulty of the question based on model solvability, and selects questions using the average pass rate of the model across samples for RL training. On Qwen-2.5-7B, our method achieves a 29.2% absolute improvement over the base model on GSM8K (+2.9 pp vs. instruction-tuned, +6.6 pp vs. Self-Instruct), 8.7% on MATH, 13.1% on GPQA (+7.0 pp vs. SynthLLM), 8.9% on MedQA, 17.7% on CQA (law) and 13.7% on CFA (finance). It surpasses supervised fine-tuning under the same data budget and nearly matches RL with full human data across datasets (e.g., +17.2 pp on GSM8K). Adding 100 human demonstrations improves the performance of GSM8K only by 0.4 pp, showing a limited added value. By reducing human data annotation, Synthetic Data RL enables scalable and efficient RL-based model adaptation. Code and demos are available at https://github.com/gydpku/Data_Synthesis_RL/.

cs.CL↗