arXiv subjects

Shiqi Hu

Publications and source records attributed to Shiqi Hu.

4 recordsLinked to original sources

arXiv2026

Critical Role of Hydrogen in Unconventional Superconductors: The Case of Hydrogenated FeSe Layers

Hydrogenation is known to tune superconductivity in a wide range of materials. While its microscopic role has been clarified in phonon-mediated superconductors such as hydrogenated MgB2, LaH10, and H3S, much less is known for hydrogenated cuprates and iron-based superconductors, where even the underlying structural motifs remain elusive. Using hydrogenated FeSe as a prototypical example, we reveal how hydrogen affects superconductivity in the presence of strong electronic correlations: correlation-induced orbital renormalization shifts hydrogen-derived spectral weight from the high-energy region toward the Fermi surface (FS), remarkably enhancing the electron-phonon coupling (EPC). We predict a structurally stable FeSeH phase where, compared to bare FeSe, hydrogen incorporation reshapes the FS topology and increases the number of channels for electron-phonon scattering, while simultaneously introducing high-frequency phonons that strengthen pairing. First-principles EPC calculations combined with dynamical mean field theory (DMFT) yield a superconducting transition temperature (Tc) exceeding 40 K. Fully anisotropic Eliashberg theory reveals a two-gap superconducting state, consistent with the gap structure experimentally observed in doped FeSe. Our findings identify correlation-enhanced EPC as a plausible microscopic mechanism for iron-based superconductivity and offer a new perspective on pairing in strongly correlated systems. In addition, this work establishes hydrogenated FeSe as a promising platform for engineering two-dimensional superconductors and superconducting quantum devices.

cond-mat.supr-con↗

arXiv2026

Ultrafast N\'eel vector switching

We predict ultrafast switching in a chiral anti-ferromagnet that occurs at femtosecond times, nearly 5 orders of magnitude faster than the torque induced nanosecond switching previously observed. The physical mechanism, quite different from that which drives slow switching, involves the creation of massive effective magnetic fields by ultrafast spin current injection. Identifying these fields as key to femtosecond rotation, we establish simple practical rules for their maximisation with wide applicability to all magnetised materials. Employing state-of-the-art time-dependent density-functional theory and using the example of chiral magnet, Mn$_3$Sn, we induce ultrafast rotation enough to drive the switching of magnetic order between the six possible non-collinear ground states. We further demonstrate the possibility of undoing this switching by subsequent injection of oppositely polarized spin current. Our findings place chiral anti-ferromagnets as a materials platform for femtosecond N\'eel-vector switching, opening a route towards the manipulation of magnetic matter at ultrafast times.

cond-mat.mtrl-sci↗

arXiv2026

OSCBench: Benchmarking Object State Change in Text-to-Video Generation

Text-to-video (T2V) generation models have made rapid progress in producing visually high-quality and temporally coherent videos. However, existing benchmarks primarily focus on perceptual quality, text-video alignment, or physical plausibility, leaving a critical aspect of action understanding largely unexplored: object state change (OSC) explicitly specified in the text prompt. OSC refers to the transformation of an object's state induced by an action, such as peeling a potato or slicing a lemon. In this paper, we introduce OSCBench, a benchmark specifically designed to assess OSC performance in T2V models. OSCBench is constructed from instructional cooking data and systematically organizes action-object interactions into regular, novel, and compositional scenarios to probe both in-distribution performance and generalization. We evaluate six representative open-source and proprietary T2V models using both human user study and multimodal large language model (MLLM)-based automatic evaluation. Our results show that, despite strong performance on semantic and scene alignment, current T2V models consistently struggle with accurate and temporally consistent object state changes, especially in novel and compositional settings. These findings position OSC as a key bottleneck in text-to-video generation and establish OSCBench as a diagnostic benchmark for advancing state-aware video generation models.

cs.CV↗

arXiv2024

Optimizing attosecond pulse generation in solids by modulating electronic dynamics with monochromatic laser field

A practical approach is proposed for efficiently generating ultrashort attosecond pulses (APs) from realistic solid-state materials, aiming to optimize pulse width effectively. By adjusting the photon energy while maintaining a constant peak electric field, this strategy modulates the peak vector potential and laser field period, thereby controlling the high harmonic cutoff energy and the time-domain emission characteristics of the harmonics. The field-driven electronic dynamics lead to a non-monotonic variation in both the intensity and duration of the generated attosecond pulses. The light field frequency can be adjusted to yield the optimal pulse. Beyond the primary demonstration with hexagonal boron nitride as a prototypical material, significant pulse width optimization has been achieved across a range of different materials. This straightforward and versatile strategy shows promise for application in solid-state materials, offering new pathways to promote high harmonic performance.

physics.optics↗