arXiv subjects

Wei Qin

Publications and source records attributed to Wei Qin.

At least 19 recordsLinked to original sources

arXiv2026

Beyond Cross-Lingual Transfer: Benchmarking Propagation Boundaries in Multilingual LLM Unlearning

Large Language Model (LLM) unlearning aims to suppress target knowledge while preserving general capabilities. In multilingual settings, unlearning must additionally propagate within its intended linguistic scope. However, existing evaluations mainly measure cross-lingual transfer and cannot distinguish insufficient from excessive propagation. We introduce CLLPU (Cross-Lingual and Language-Bound Protocol for LLM Unlearning), a multilingual benchmark that formulates this problem through two settings: common-goal forgetting, where target knowledge should be suppressed across all languages, and language-conditioned forgetting, where suppression should remain confined to a designated language. CLLPU combines goal-guided topic pairing, schema-aware relation matching, and dual-anchor multilingual translation to construct 800 matched knowledge-unit pairs and 72,000 QA instances across ten languages. Experiments with six representative methods on Llama-3.1-8B-Instruct reveal opposite failure modes: forgetting remains incomplete when universal suppression is required, yet spreads beyond the intended boundary when language-conditioned confinement is required. We further find that general multilingual utility can conceal damage to neighbor knowledge. These findings establish propagation control as a central challenge for multilingual LLM unlearning. We publicly release CLLPU together with its construction pipeline.

cs.CL↗

arXiv2026

Tuning Density and Spin Ordering of Degenerate Fermi Gases in an Optical Cavity

We investigate a spin-degenerate Fermi gas coupled to a high-finesse optical cavity, where the competition between scalar and vectorial couplings is controlled by the relative polarization angle of the pump and cavity fields. We find that the phase transition threshold is synergistically determined by the scalar-vectorial coupling weight and Pauli blocking, with the latter dictating the critical pump lattice depth required for the onset of superradiance. For a two-component Fermi gas with opposite spins, the population ratio drives two distinct types of phase transitions corresponding to real-space phase separation: continuous and discontinuous. Nevertheless, the boundary of the phase transition remains fundamentally governed by the scalar-vectorial coupling competition. We clarify the impact of the relative polarization angle on phase transitions of the system; these results also apply to bosonic systems. Our results provide valuable theoretical insights for future experimental realizations.

cond-mat.quant-gas↗

arXiv2026

Qubit Readout via State-Dependent Radiative Linewidths

Fast qubit readout conventionally encodes state information in a dispersive frequency shift. Here we formulate a linewidth-encoded quantum non-demolition measurement channel in which the qubit state enters the external radiative amplitude, equivalently a state-dependent Lindblad jump operator. Starting from an empty cavity, we show analytically that this dissipative channel imprints state information on the output field at $O(t)$, whereas standard dispersive readout starts at $O(t^2)$ because it requires intracavity buildup and conditional phase accumulation. This short-time scaling produces faster matched-filter signal-to-noise ratio accumulation and persists in finite-resource comparisons, including photon-number limits, external-linewidth budgets, cavity depletion, and pulse-optimized dispersive baselines. We further outline an auxiliary-mode route that converts a qubit-state-dependent auxiliary susceptibility into a state-dependent linewidth. These results identify engineered dissipation as an information-carrying resource for fast quantum non-demolition readout.

quant-ph↗

arXiv2026

CompassDPO: Dynamics-Controlled Direct Preference Optimization for Robust Safety Alignment

Direct Preference Optimization (DPO) has become a standard framework for safety alignment, but its reliance on pairwise preference updates makes training sensitive to imperfect supervision. Existing robust DPO methods often address this sensitivity through global loss corrections or external data-level interventions, while largely overlooking how unreliable comparisons distort batch-level optimization dynamics. We propose CompassDPO, a reward-free DPO framework that stabilizes preference optimization through dynamics control. Using the implicit DPO reward margin as a training-time compass, CompassDPO regulates sample influence along two complementary axes: update direction and update magnitude. For directional control, it applies sparse, budgeted, and warm-up delayed loss mixing to attenuate update components that conflict with the emerging preference direction. For magnitude control, it adaptively soft-winsorizes high-loss tail contributions, reducing tail dominance while preserving useful gradients from hard examples. Both mechanisms use only signals available during standard DPO training and require no external reward model or additional supervision. Experiments on PKU-SafeRLHF across four backbones and multiple out-of-distribution safety benchmarks show that CompassDPO consistently improves robustness over vanilla DPO and strong DPO-family baselines, especially under controlled label-flip noise. Code is available at https://anonymous.4open.science/r/CompassDPO-4D00

cs.LG↗

arXiv2026

Phase transition, phase separation and mode softening of a two-component Bose-Einstein condensate in an optical cavity

We investigate the superradiant phase transition in a two-component Bose-Einstein condensate with distinct atomic detunings, confined in an optical cavity and driven by a transverse pump laser. By combining perturbation theory and numerical simulations, we demonstrate that the phase transition is dominated by the red-detuned component, resulting in a phase diagram completely different from that of a single-component case under blue-detuned condition. The system exhibits spontaneous phase separation between the two components, manifested as alternating stripe patterns in the normal phase and distinct Bragg gratings in the superradiant phase. Furthermore, the Bogoliubov excitation spectrum reveals roton-type mode softening, indicating that the phase transition also corresponds to the superfluid-to-lattice supersolid transition. Our findings provide insights into the interplay between atomic detunings and collective quantum many-body phenomena, offering potential applications in quantum simulation and optical switching technologies.

cond-mat.quant-gas↗

arXiv2025

Debate over Mixed-knowledge: A Robust Multi-Agent Reasoning Framework for Incomplete Knowledge Graph Question Answering

Knowledge Graph Question Answering (KGQA) aims to improve factual accuracy by leveraging structured knowledge. However, real-world Knowledge Graphs (KGs) are often incomplete, leading to the problem of Incomplete KGQA (IKGQA). A common solution is to incorporate external data to fill knowledge gaps, but existing methods lack the capacity to adaptively and contextually fuse multiple sources, failing to fully exploit their complementary strengths. To this end, we propose Debate over Mixed-knowledge (DoM), a novel framework that enables dynamic integration of structured and unstructured knowledge for IKGQA. Built upon the Multi-Agent Debate paradigm, DoM assigns specialized agents to perform inference over knowledge graphs and external texts separately, and coordinates their outputs through iterative interaction. It decomposes the input question into sub-questions, retrieves evidence via dual agents (KG and Retrieval-Augmented Generation, RAG), and employs a judge agent to evaluate and aggregate intermediate answers. This collaboration exploits knowledge complementarity and enhances robustness to KG incompleteness. In addition, existing IKGQA datasets simulate incompleteness by randomly removing triples, failing to capture the irregular and unpredictable nature of real-world knowledge incompleteness. To address this, we introduce a new dataset, Incomplete Knowledge Graph WebQuestions, constructed by leveraging real-world knowledge updates. These updates reflect knowledge beyond the static scope of KGs, yielding a more realistic and challenging benchmark. Through extensive experiments, we show that DoM consistently outperforms state-of-the-art baselines.

cs.AI↗

arXiv2025

GeoLink: Empowering Remote Sensing Foundation Model with OpenStreetMap Data

Integrating ground-level geospatial data with rich geographic context, like OpenStreetMap (OSM), into remote sensing (RS) foundation models (FMs) is essential for advancing geospatial intelligence and supporting a broad spectrum of tasks. However, modality gap between RS and OSM data, including differences in data structure, content, and spatial granularity, makes effective synergy highly challenging, and most existing RS FMs focus on imagery alone. To this end, this study presents GeoLink, a multimodal framework that leverages OSM data to enhance RS FM during both the pretraining and downstream task stages. Specifically, GeoLink enhances RS self-supervised pretraining using multi-granularity learning signals derived from OSM data, guided by cross-modal spatial correlations for information interaction and collaboration. It also introduces image mask-reconstruction to enable sparse input for efficient pretraining. For downstream tasks, GeoLink generates both unimodal and multimodal fine-grained encodings to support a wide range of applications, from common RS interpretation tasks like land cover classification to more comprehensive geographic tasks like urban function zone mapping. Extensive experiments show that incorporating OSM data during pretraining enhances the performance of the RS image encoder, while fusing RS and OSM data in downstream tasks improves the FM's adaptability to complex geographic scenarios. These results underscore the potential of multimodal synergy in advancing high-level geospatial artificial intelligence. Moreover, we find that spatial correlation plays a crucial role in enabling effective multimodal geospatial data integration. Code, checkpoints, and using examples are released at https://github.com/bailubin/GeoLink_NeurIPS2025

cs.CV↗

arXiv2025

Signature of chiral superconducting order parameter evidenced in mesoscopic superconductors

Chiral superconductivity is a novel superconducting phase characterized by order parameters that break the time-reversal symmetry, endowing the state with a definite handedness. Unlike conventional superconductors, the Cooper pairs in a chiral superconductor carry nonzero orbital angular momentum. Through coupling with an external magnetic field, the finite angular momentum of the Cooper pair modulates the temperature-magnetic field phase boundary in a distinctive way, which could serve as an experimental signature of the chiral superconducting state. Here we demonstrate that the chiral signature can be detected in mesoscopic superconducting rings of $\beta$-Bi$_2$Pd, manifesting as a linear-in-field modulation of the critical temperature in the Little-Parks effect. Our findings establish a new experimental method for detecting the chiral superconductivity.

cond-mat.supr-con↗

arXiv2025

Tight Cache Contention Analysis for WCET Estimation on Multicore Systems

WCET (Worst-Case Execution Time) estimation on multicore architecture is particularly challenging mainly due to the complex accesses over cache shared by multiple cores. Existing analysis identifies possible contentions between parallel tasks by leveraging the partial order of the tasks or their program regions. Unfortunately, they overestimate the number of cache misses caused by a remote block access without considering the actual cache state and the number of accesses. This paper reports a new analysis for inter-core cache contention. Based on the order of program regions in a task, we first identify memory references that could be affected if a remote access occurs in a region. Afterwards, a fine-grained contention analysis is constructed that computes the number of cache misses based on the access quantity of local and remote blocks. We demonstrate that the overall inter-core cache interference of a task can be obtained via dynamic programming. Experiments show that compared to existing methods, the proposed analysis reduces inter-core cache interference and WCET estimations by 52.31% and 8.94% on average, without significantly increasing computation overhead.

cs.SE↗

arXiv2025

Pressure-Driven Moir\'e Potential Enhancement and Tertiary Gap Opening in Graphene/h-BN Heterostructure

Moir\'e superlattices enable engineering of correlated quantum states through tunable periodic potentials, where twist angle controls periodicity but dynamic potential strength modulation remains challenging. Here, we develop a high-pressure quantum transport technique for van der Waals heterostructures, achieving the ultimate pressure limit (~9 GPa) in encapsulated moir\'e devices. In aligned graphene/h-BN, we demonstrate that pressure induces a substantial enhancement of the moir\'e potential strength, evidenced by the suppression of the first valence bandwidth and the near-doubling of the primary band gap. Moreover, we report the first observation of a tertiary gap emerging above 6.4 GPa, verifying theoretical predictions. Our results establish hydrostatic pressure as a universal parameter to reshape moir\'e band structures. By enabling quantum transport studies at previously inaccessible pressure regimes, this Letter expands the accessible parameter space for exploring correlated phases in moir\'e systems.

cond-mat.str-el↗

arXiv2025

Three-dimensional numerical study on hydrogen bubble growth at electrode

Three-dimensional direct numerical simulation of electrolysis is applied to investigate the growth and detachment of bubbles at electrodes. The moving gas-liquid interface is modeled employing the VOF-based method. To ensure the accuracy of the simulations, a mesh-independence study has been performed. The simulations include the growth phase of the bubbles, followed by their detachment from the electrode surface, and the results are validated with analytical models and experimental data. The bubble growth is diffusion-controlled, leading to the scaling $R\propto t^{1/2}$, but our simulation overpredicts the growth exponent during the initial stage. We further demonstrate that the number of nucleation sites significantly affects gas transport, as quantified by the Sherwood number. The influences of contact angle and nucleation site on bubble detachment are also examined. The predicted detachment radius varies linearly with contact angle, consistent with Fritz's linear relation between the volume-equivalent radius and contact angle, confirming that the surface tension is the dominant attachment force. Finally, as the nucleation sites increase, the induced bubble coalescence accelerates the bubble detachment. Taken together, these findings give us valuable insights into improving gas bubble removal and enhancing overall electrolysis efficiency.

physics.flu-dyn↗

arXiv2025

Quantum phase transition driven by competing intralayer and interlayer hopping in bilayer nickelates

Bilayer nickelates exhibit high-temperature superconductivity under proper hydrostatic pressure or epitaxial strain, signifying the emergence of quantum phase transitions whose physical mechanisms remain unclear. Using a minimal bilayer Hubbard model incorporating only the Ni-$d_{3z^2-r^2}$ orbitals, we demonstrate that a phase transition naturally arises from tuning the ratio of intralayer to interlayer hopping amplitudes. The transition point separates regimes with a rich interplay between superconducting and density-wave orders. In the regime of weaker intralayer hopping, the ground state is characterized by quasi-long-range spin-density-wave order. As the intralayer hopping increases, the system undergoes a transition marked by the opening of a finite spin gap and the disappearance of spin-density-wave order. Meanwhile, superconductivity is dramatically enhanced, accompanied by the emergence of quasi-long-range charge-density-wave order, indicating that the system enters Luther-Emery phase. This quantum phase transition, driven by the competition between intralayer and interlayer hopping, provides a plausible microscopic explanation for the experimentally observed correlation between the superconducting transition temperature and ratio of out-of-plane to in-plane lattice constants. Our findings reveal a possible link between the suppression of spin-density-wave order and the prominence of superconducting order, which may assist future efforts to optimize experimental conditions for further enhancing superconductivity in bilayer nickelates.

cond-mat.str-el↗

arXiv2025

Prediction of High-Temperature Half Quantum Anomalous Hall Effect in a Semi-magnetic Topological Insulator of MnBi$_2$Te$_4$/Sb$_2$Te$_3$

The classic Thouless-Kohmoto-Nightingale-Nijs theorem dictates that a single electron band of a lattice can only harbor an integer quantum Hall conductance as a multiple of e^2/2h, while recent studies have pointed to the emergence of half quantum anomalous Hall (HQAH) effect, though the underlying microscopic mechanisms remain controversial. Here we propose an ideal platform of MnBi$_2$Te$_4$/Sb$_2$Te$_3$ that allows not only to realize the HQAH effect at much higher temperatures, but also to critically assess the different contributions of the gapped and gapless Dirac bands. We first show that the top surface bands of the Sb$_2$Te$_3$ film become gapped, while the bottom surface bands remain gapless due to proximity coupling with the MnBi$_2$Te$_4$ overlayer. Next we show that such a semi-magnetic topological insulator harbors the HQAH effect at ~20 K, with Cr doping enhancing it to as high as 67 K, driven by large magnetic anisotropy and strong magnetic coupling constants that raise the Curie temperature. Our detailed Berry curvature analysis further helps to reveal that, whereas the gapped surface bands can contribute to the Hall conductance when the chemical potential is tuned to overlap with the bands, these bands have no net contribution when the chemical potential is in the gapped region, leaving the gapless bands to be the sole contributor to the HQAH conductance. Counterintuitively, the part of the gapless bands within the gapped region of the top surface bands have no net contribution, thereby ensuring the plateau nature of the Hall conductance.

cond-mat.mes-hall↗

arXiv2025

Ferroelectrically Switchable Half-Quantized Hall Effect

Integrating ferroelectricity, antiferromagnetism, and topological quantum transport within a single material is rare, but crucial for developing next-generation quantum devices. Here, we propose a multiferroic heterostructure consisting of an antiferromagnetic MnBi$_2$Te$_4$ bilayer and an Sb$_2$Te$_3$ film is able to harbor the half-quantized Hall (HQH) effect with a ferroelectrically switchable Hall conductivity of $e^2/2h$. We first show that, in the energetically stable configuration, the antiferromagnetic MnBi$_2$Te$_4$ bilayer opens a gap in the top surface bands of Sb$_2$Te$_3$ through proximity effect, while its bottom surface bands remain gapless; consequently, HQH conductivity of $e^2/2h$ can be sustained clockwise or counterclockwise depending on antiferromagnetic configuration of the MnBi$_2$Te$_4$. Remarkably, when applying interlayer sliding within the MnBi$_2$Te$_4$ bilayer, its electric polarization direction associated with parity-time reversal symmetry breaking is reversed, accompanied by a reversal of the HQH conductivity. The proposed approach offers a powerful route to control topological quantum transport in antiferromagnetic materials by ferroelectricity.

cond-mat.mes-hall↗

arXiv2025

Topological Flat Minibands and Fractional Chern Insulators in Rashba Systems with Tunable Superlattice Potentials

Fractional Chern insulators are interaction-driven topological states that realize the lattice analogue of the fractional quantum Hall effect without external magnetic fields. Here we propose a programmable platform for engineering topological flat bands by combining a Zeeman field with a tunable electrostatic superlattice potential in a Rashba spin-orbit-coupled thin film. The Zeeman field generates a nearly flat segment of the Rashba band, which is isolated by the superlattice potential into a flat band with Chern number $\mathcal{C}=1$. Using many-body exact diagonalization, we show that this band supports fractional Chern insulators at filling factors $n = 1/3$ and $2/3$, both of which remain robust over broad parameter ranges. We further identify realistic material candidates and the corresponding device conditions that enable experimental realization. These results establish a versatile and experimentally accessible platform for engineering topological flat bands and exploring correlated topological phases.

cond-mat.mes-hall↗

arXiv2025

Quantum gyroscope based on the cavity magnomechanical system

High-precision rotational angle measurement in noise-prone environments holds critical impor tance in aerospace engineering, military navigation, and related domains. In this paper, we propose a quantum gyroscope scheme based on a cavity magnomechanical system, which enables high precision rotation angle detection by harnessing hybrid light-magnon interactions. Central to this framework is the employment of a two-mode squeezed coherent state, generated via parametric coupling of dual quantized optical fields with collective spin excitations (magnons), serving as the quantum metrological probe. We demonstrate that this scheme can significantly reduce quantum noise to levels far below the shot-noise limit. Furthermore, in the non-Markovian case, the per formance of the quantum gyroscope in a dissipative environment does not deteriorate over time, provided that the environmental spectral density satisfies certain conditions. These findings provide critical insights for advancing miniaturized quantum gyroscopes with sub-microradian precision, addressing long-standing challenges in inertial navigation systems under strong ambient noise.

quant-ph↗

arXiv2025

Distributed Graph Neural Network Inference With Just-In-Time Compilation For Industry-Scale Graphs

Graph neural networks (GNNs) have delivered remarkable results in various fields. However, the rapid increase in the scale of graph data has introduced significant performance bottlenecks for GNN inference. Both computational complexity and memory usage have risen dramatically, with memory becoming a critical limitation. Although graph sampling-based subgraph learning methods can help mitigate computational and memory demands, they come with drawbacks such as information loss and high redundant computation among subgraphs. This paper introduces an innovative processing paradgim for distributed graph learning that abstracts GNNs with a new set of programming interfaces and leverages Just-In-Time (JIT) compilation technology to its full potential. This paradigm enables GNNs to highly exploit the computational resources of distributed clusters by eliminating the drawbacks of subgraph learning methods, leading to a more efficient inference process. Our experimental results demonstrate that on industry-scale graphs of up to \textbf{500 million nodes and 22.4 billion edges}, our method can produce a performance boost of up to \textbf{27.4 times}.

cs.LG↗

arXiv2025

Theoretical Exploration of Phase Transitions in a Cavity-BEC System with Two Crossed Optical Pumps

We consider a Bose-Einstein condensation (BEC) inside an optical cavity and two crossed coherent pump fields. We determine the phase boundary separating the normal superfluid phase and the superradiance phase, perturbatively. In the regime of negative cavity detuning, we map out the phase diagrams both for an attractive and a repulsive optical lattice. It turns out that the situation is quite different in two cases. Specifically, in the case of attractive lattice, if a system is in the superradiant phase with one pump laser, adding another pump does not drive the system out of the superradiance phase. While for the repulsive lattice, increasing another pump potential have suppressive effects on the superradiance. We also find that, in the case of attractive lattice, equally increasing two pump lattice potentials can induce a transition from the normal phase to the superradiance phase. In stark contrast, for the repulsive lattice, the system will remain in the normal phase as the pump depths are tuned within a wide range, independent of the cavity detuning and the decay rate.

cond-mat.quant-gas↗