arXiv subjects

Fei Zhou

Publications and source records attributed to Fei Zhou.

At least 19 recordsLinked to original sources

arXiv2026

S$^2$-RL: Sample-Set Dual Reinforcement Learning for Generative Semantic Segmentation Dataset Distillation

While dataset distillation has witnessed significant advances in image classification, its extension to dense prediction tasks such as semantic segmentation is plagued by two core bottlenecks: bi-level optimization-based approaches suffer from prohibitive computational costs stemming from pixel-wise gradient unrolling, whereas proxy-based generative methods merely infuse visual textures into fixed semantic masks derived from real data, thus failing to condense rich semantic knowledge into a compact set of informative, novel scene generations. To address both bottlenecks simultaneously, we propose a novel Sample-Set Reinforcement Learning (S$^2$-RL) framework for generative semantic segmentation dataset distillation (SSDD). Specifically, S$^2$-RL formulates SSDD as a diffusion model based text-to-image generation paradigm, enabling flexible generation of images with informative, novel semantic distributions via a concise text prompt. Furthermore, we fine-tune a diffusion policy via Group Relative Policy Optimization (GRPO), leveraging a sample-set dual reward paradigm: for individual samples, it enforces semantic alignment with input text prompts and intra-class feature diversity; for sample sets, it maximizes inter-sample semantic distribution diversity by solving a contextual multi-armed bandit problem. These design choices enable S$^2$-RL to distill large-scale semantic segmentation datasets into a compact set of generated samples that encapsulate diverse, informative semantic knowledge without the need of pixel-wise gradient unrolling. Extensive experiments on the ADE20K and COCO datasets demonstrate that S$^2$-RL achieves substantial and consistent improvements over state-of-the-art baselines, establishing a new benchmark for SSDD.

cs.CV↗

arXiv2026

From Stages to States: Unified Iterative Reasoning for MLLMs

Multimodal Chain-of-Thought (MCoT) substantially enhances the complex reasoning capabilities of Multimodal Large Language Models (MLLMs) by explicitly modeling intermediate reasoning processes. However, most learning-based MCoT methods follow a stage-centric paradigm, where different reasoning behaviors are assigned to predefined stages and optimized separately. This results in fragmented reasoning and a training-inference reasoning-state gap. To address these limitations, we propose ReState, a unified iterative reasoning framework that reformulates multimodal reasoning from a stage-driven pipeline into a state-driven dynamic process. It comprises two complementary components: Unified Reasoning State Modeling (URSM) and Closed-Loop Reasoning-State Learning (CRSL). URSM unifies reasoning construction, correct reasoning preservation, and erroneous reasoning correction as reasoning-state transitions within a shared state space, enabling these heterogeneous reasoning behaviors to be learned jointly. CRSL further incorporates self-generated historical reasoning states during training and recursively updates the current state during inference, thereby mitigating the reasoning-state gap and enabling iterative self-refinement without requiring an additional reflection module, critic, or independent correction stage. Furthermore, we construct VersaCoT-10K, a high-quality CoT dataset spanning eight visual reasoning domains, which provides diverse supervision for multi-domain reasoning and reasoning-state refinement. Extensive experiments demonstrate that ReState exhibits strong adaptability across different MLLM backbones and achieves strong performance on multiple public reasoning benchmarks, supporting the shift from stage-centric reasoning toward state-centric reasoning. Our data and code are available at https://github.com/YuChuang1205/ReState

cs.AI↗

arXiv2026

MOSAIC: Addressing Space for Local, Steerable, and Scalable Atomistic Generation

Generative models are opening new possibilities in molecular discovery and materials science. Realizing this potential, however, requires both local control over atomic placement and the ability to assemble large, heterogeneous structures. We introduce Mosaic, a hybrid mesh-particle representation that partitions space into anchored regions containing atom slots. Each slot encodes occupancy, atomic species, and a continuous displacement from its anchor, providing persistent spatial addresses for local conditioning, hierarchical generation, and scalable assembly. Because Mosaic defines a representation rather than a model, it can be paired with different generative backbones. With the molecular backbone held fixed, its local support and coordinate prior improve molecular connectivity and validity, the resulting model samples efficiently across molecular sizes and enables fine-grained spatial control. At larger material scales, the same representation enables generation of atomistic polycrystals conditioned on coarse spatial layouts. Most notably, locality turns large-scale generation into a bounded assembly problem, enabling structure generation with more than one million atoms. Together, this establishes local spatial addressing as a practical foundation for steerable generation of matter across scales, from individual molecules to large-scale structures of broader engineering significance.

cond-mat.mtrl-sci↗

arXiv2026

Topologically shadowed quantum criticality: A non-compact conformal manifold

We propose a family of topological quantum critical points (tQCPs) separating non-invertible chiral topological orders in $(2+1)$ dimensions. We conjecture that these tQCPs can be captured by a family of scale-invariant field theories forming a non-compact scale-invariant manifold. A central feature of our proposal is topological shadowing: the underlying critical theory is rigorously constrained by the global topological data of the two adjacent gapped phases. These theories can be further projected into quantum field theories with universal non-local structures. The non-local coupling, however, controls dynamical observables along the scale-invariant manifold, including the VEV of Wilson loops and their exchange phase around the two non-contractible cycles of a torus. Despite the non-locality, our renormalization group calculations (up to two-loop order) strongly suggest that the theory shall maintain exact scale invariance. This supports, without supersymmetry, a continuous family of scale-invariant fixed points, which are further substantiated from a holographic perspective.

cond-mat.str-el↗

arXiv2026

Holographic Representations of Topological Quantum Criticality: Emergent Symmetry Approach around the Bott Clock

In this article, we apply the idea of emergent symmetries to construct holographic representations of a broad class of topological quantum critical points (tQCPs) that appear in fermionic classes in the Bott clock. We show explicitly that around the Bott clock, an emergent symmetry $U_{EM}$ exists at a tQCP between different symmetry protected gapped phases with protecting symmetry $G_p$ in $d$ spatial dimensions. This emergent symmetry can be used to construct a $(d+1)$ spatial dimensional lattice model with a properly upgraded symmetry such as $U_{EM}\times G_p$ or $U_{EM} \rtimes G_p$, etc. By doubling the degrees of freedom of the adjacent topological class in $d$-dimensions, we successfully show that the $(d+1)$-dimensional lattice models constructed in this way have the desired enlarged symmetry groups that precisely belong to the adjacent topological classes, counterclockwise around the Bott clock. Furthermore, $d$-dimensional boundaries of these $(d+1)$ dimensional lattice models of gapped topological classes are shown to exhibit identical infrared dynamics as the corresponding tQCPs in the $d$-dimensional adjacent topological phases in the Bott clock, with lower symmetries.

cond-mat.str-el↗

arXiv2026

Extracting Atomic Environments for Machine Learning Interatomic Potentials

In order to appropriately capture large-scale material features and emergent phenomena via atomistic simulations, such as Molecular Dynamics (MD), the system scale can range up to hundreds of millions of atoms. However, the force-field models that drive those simulations are generally trained with Density Functional Theory (DFT) reference data, limited to relatively small configurations on the order of 100s or 1000s of atoms. To compute DFT forces on atoms in regions of interest, for example for active-learning or on-the-fly training of interatomic potentials, one needs to extract a small set of atoms from the larger simulation box, and typically work with periodic boundary conditions for DFT. However, methods to select the shape and size of this extracted set of atoms, as well as to generate a potentially necessary passivating envelope, have not been systematically analyzed. In this work, we benchmark several techniques, including a generative diffusion-based artificial intelligence (AI) approach, for extracting atomic environments from large, bulk configurations and embedding them into smaller configurations suitable for DFT calculations with periodic boundary conditions. We test with a diverse set of material systems, which includes amorphous $\mathrm{SiO_2}$, Ta with screw dislocations, and molten C. We demonstrated a notably simple procedure, a method we refer to as deletions, yields superior performance over an array of alternative extraction methods.

cond-mat.mtrl-sci↗

arXiv2026

Single-atom sensor for low-frequency electric field

Precision measurement of low-frequency electric field (LFEF) signals with frequency from 30 kHz to 300 kHz is crucial for advancing both fundamental science and practical applications, owing to their unique frequency regime. For conventional electromagnetic antennas, the long wavelength (i.e., several kilometers) of the LFEF leads to a severe size constraint that efficient radiation becomes challenging to achieve when the antenna size is much smaller than the long wavelength of the LFEF signals, which in turn results in a reduction of measurement sensitivity and compromises antenna's performance. By exploiting the high intrinsic sensitivity of cold trapped ions to weak alternating electric signals via Coulomb interaction, we demonstrate a single-ion phonon laser sensor acted by an injection-locked 40Ca+ ion confined in a surface-electrode trap. Combining the beat frequency technique with the injection-locked phonon laser oscillation, we demonstrate a practical and efficient approach for simultaneous extraction of the frequency, phase, and amplitude from a single measurement, without the need for sideband cooling. This approach achieves precision detection for LFEF signals with the sensitivity of 404 uV/(m * Hz1/2) and the detection limit of 61.5 uV/m. Besides, this approach also shows remarkable robustness against noise. Our study helps realizing practical single-atom sensors in the low-frequency regime, opening avenues for applications in subsurface communication, precision metrology, mass spectrometry, and biomedical monitoring.

quant-ph↗

arXiv2026

Enhanced detection of electric field signals via squeezing-induced stochastic resonance

Stochastic resonance (SR) could amplify weak electric-field signals in nonlinear systems by means of the externally injected noises. Here we propose and experimentally demonstrate a modified SR method, termed squeezing-induced SR, implemented in the system involving a trapped ion behaving as a Duffing oscillator. We find that squeezing the phase noise of the oscillator results in amplified fluctuation of the corresponding amplitude, which helps achieve the SR. Since no auxiliary noise source is needed, the squeezing-induced SR may enhance the signal-to-noise ratio by 4.28 $\pm$ 0.39 dB compared to the conventional noise-induced SR under identical conditions of the electric-field detection. This technique offers a promising approach for developing atomic ion sensors for detecting weak electric-field signals.

physics.atom-ph↗

arXiv2026

UniTriGen: Unified Triplet Generation of Aligned Visible-Infrared-Label for Few-Shot RGB-T Semantic Segmentation

RGB-T semantic segmentation requires strictly aligned VIS-IR-Label triplets; however, such aligned triplet data are often scarce in real-world scenarios. Existing generative augmentation methods usually adopt cascaded generation paradigms, decomposing joint triplet generation into local conditional processes. As a result, consistency among VIS, IR, and Label in spatial structure, semantic content, and cross-modal details cannot be reliably maintained. To address this issue, we propose UniTriGen, a unified triplet generation framework that directly generates spatially aligned, semantically consistent, and modality complementary VIS-IR-Label triplets under the guidance of text prompts. UniTriGen first introduces a unified triplet generation mechanism, where VIS, IR, and Label are jointly encoded into a shared latent space and modeled with a diffusion process to enforce global cross-modal consistency. Lightweight modality-specific residual adapters are further integrated into this mechanism to accommodate modality-specific imaging characteristics and output formats. To mitigate generation bias caused by imbalanced scene and class distributions in limited paired triplets, UniTriGen also employs a scene-balanced and class-aware few-shot sampling strategy, which induces a more balanced sampling distribution and enhances the scene and class diversity of generated triplets. Experiments show that UniTriGen generates high-quality aligned triplets from limited real paired data, thereby achieving consistent performance improvements across various RGB-T semantic segmentation models.

cs.CV↗

arXiv2026

Equivalence class of Emergent Single Weyl fermion lattice models in 3 dimensions: gapless superconductors and superfluids versus chiral fermions

In this article, we put forward a practical but generic approach towards constructing a large family of $(3+1)$ dimension lattice models which can naturally lead to a single Weyl cone in the infrared (IR) limit. Our proposal relies on spontaneous charge $U(1)$ symmetry breaking to evade the usual no-go theorem of a single Weyl cone in a 3d lattice. We have explored three concrete paths in this approach, all involving fermionic topological symmetry protected states (SPTs). Path a) is to push a gapped SPT in a 3d lattice with time-reversal symmetry (or $T$-symmetry) to a gapless topological quantum critical point (tQCP) which involves a minimum change of topologies,i.e. $δN_w=2$ where $δN_w$ is the change of winding numbers across the tQCP. Path b) is to peal off excessive degrees of freedom in the gapped SPT via applying $T$-symmetry breaking fields which naturally result in a pair of gapless nodal points of real fermions. Path c) is a hybrid of a) and b) where tQCPs, with $δN_w \geq 2$, are further subject to time-reversal-symmetry breaking actions. In the infrared limit, all the lattice models with single Weyl fermions studied here are isomorphic to either a tQCP in a DIII class topological superconductor with a protecting $T$-symmetry, or its dual, a $T$-symmetry breaking superconducting nodal point phase, and therefore form an equivalent class. For a generic $T$-symmetric tQCP along Path a), the conserved-charge operators span a six-dimensional linear space while for a $T$-symmetry breaking gapless state along Path b), c), charge operators typically span a two-dimensional linear space instead. Finally, we pinpoint connections between three spatial dimensional lattice chiral fermion models and gapless real fermions that can naturally appear in superfluids or superconductors studied previously.

hep-th↗

arXiv2026

A probabilistic framework for crystal structure denoising, phase classification, and order parameters

Atomistic simulations generate large volumes of noisy structural data, yet extracting phase labels and continuous order parameters (OPs) in a robust and general manner remains challenging. Existing tools are often specialized to a limited set of prototypes and split thermal-noise removal, phase classification, and OP construction into separate steps. Here we present a unified probabilistic framework for analyzing noisy atomic configurations with respect to known crystal prototypes. The model predicts per-atom, per-prototype logits and aggregates them into a scalar log-probability (logP) landscape over atomic coordinates. Its gradient defines a conservative denoising field, while the logits provide local phase labels, prototype-resolved OPs, and ambiguity measures through logit margins. We train on AFLOW-mapped crystalline structures from the Materials Project with synthetic positional and elastic perturbations, then test extrapolation to stronger noise, finite-temperature disorder, point defects, water--ice coexistence, binary polymorphs, and shock-compressed Ti. A single differentiable scalar model recovers prototype identity after denoising, tracks smooth transformations such as Bain and Burgers paths, and exposes low-confidence regions near defects and phase boundaries. This provides an integrated and extensible tool for analyzing complex atomistic simulations.

cond-mat.mtrl-sci↗

arXiv2026

Learning noisy phase transition dynamics from stochastic partial differential equations

The non-equilibrium dynamics of mesoscale phase transitions are fundamentally shaped by thermal fluctuations, which not only seed instabilities but actively control kinetic pathways, including rare barrier-crossing events such as nucleation that are entirely inaccessible to deterministic models. Machine-learning surrogates for such systems must therefore represent stochasticity explicitly, enforce conservation laws by construction, and expose physically interpretable structure. We develop physics-aware surrogate models for the stochastic Cahn-Hilliard equation in 3D that satisfy all three requirements simultaneously. The key innovation is to parameterize the surrogate at the level of inter-cell fluxes, decomposing each flux into a deterministic mobility-weighted chemical-potential gradient and a learnable noise amplitude. This design guarantees exact mass conservation at every step and adds physical fluctuations to inter-cell mass transport. A learnable free energy functional provides thermodynamic interpretability, validated by independent recovery of the bulk double-well landscape, interfacial excess energy, and curvature-independent interfacial tension. Tests demonstrate accurate reproduction of ensemble statistics and noise-accelerated coarsening, with generalization to spatial domains 64 times larger in volume and temporal horizons 160x longer than those seen during training. Critically, the stochastic surrogate captures thermally activated nucleation in the metastable regime, a qualitative capability that no deterministic surrogate can provide regardless of training, thus establishing flux-level stochasticity as an architectural necessity rather than an optional enhancement.

physics.comp-ph↗

arXiv2026

Scaling Kinetic Monte-Carlo Simulations of Grain Growth with Combined Convolutional and Graph Neural Networks

Graph neural networks (GNN) have emerged as a promising machine learning method for microstructure simulations such as grain growth. However, accurate modeling of realistic grain boundary networks requires large simulation cells, which GNN has difficulty scaling up to. To alleviate the computational costs and memory footprint of GNN, we propose a hybrid architecture combining a convolutional neural network (CNN) based bijective autoencoder to compress the spatial dimensions, and a GNN that evolves the microstructure in the latent space of reduced spatial sizes. Our results demonstrate that the new design significantly reduces computational costs with using fewer message passing layer (from 12 down to 3) compared with GNN alone. The reduction in computational cost becomes more pronounced as the spatial size increases, indicating strong computational scalability. For the largest mesh evaluated (160^3), our method reduces memory usage and runtime in inference by 117x and 115x, respectively, compared with GNN-only baseline. More importantly, it shows higher accuracy and stronger spatiotemporal capability than the GNN-only baseline, especially in long-term testing. Such combination of scalability and accuracy is essential for simulating realistic material microstructures over extended time scales. The improvements can be attributed to the bijective autoencoder's ability to compress information losslessly from spatial domain into a high dimensional feature space, thereby producing more expressive latent features for the GNN to learn from, while also contributing its own spatiotemporal modeling capability. The training was optimized to learn from the stochastic Potts Monte Carlo method. Our findings provide a highly scalable approach for simulating grain growth.

cs.LG↗

arXiv2026

Vision-Language Model Purified Semi-Supervised Semantic Segmentation for Remote Sensing Images

The semi-supervised semantic segmentation (S4) can learn rich visual knowledge from low-cost unlabeled images. However, traditional S4 architectures all face the challenge of low-quality pseudo-labels, especially for the teacher-student framework.We propose a novel SemiEarth model that introduces vision-language models (VLMs) to address the S4 issues for the remote sensing (RS) domain. Specifically, we invent a VLM pseudo-label purifying (VLM-PP) structure to purify the teacher network's pseudo-labels, achieving substantial improvements. Especially in multi-class boundary regions of RS images, the VLM-PP module can significantly improve the quality of pseudo-labels generated by the teacher, thereby correctly guiding the student model's learning. Moreover, since VLM-PP equips VLMs with open-world capabilities and is independent of the S4 architecture, it can correct mispredicted categories in low-confidence pseudo-labels whenever a discrepancy arises between its prediction and the pseudo-label. We conducted extensive experiments on multiple RS datasets, which demonstrate that our SemiEarth achieves SOTA performance. More importantly, unlike previous SOTA RS S4 methods, our model not only achieves excellent performance but also offers good interpretability. The code is released at https://github.com/wangshanwen001/SemiEarth.

cs.CV↗

arXiv2025

Towards Efficient and Accurate Spiking Neural Networks via Adaptive Bit Allocation

Multi-bit spiking neural networks (SNNs) have recently become a heated research spot, pursuing energy-efficient and high-accurate AI. However, with more bits involved, the associated memory and computation demands escalate to the point where the performance improvements become disproportionate. Based on the insight that different layers demonstrate different importance and extra bits could be wasted and interfering, this paper presents an adaptive bit allocation strategy for direct-trained SNNs, achieving fine-grained layer-wise allocation of memory and computation resources. Thus, SNN's efficiency and accuracy can be improved. Specifically, we parametrize the temporal lengths and the bit widths of weights and spikes, and make them learnable and controllable through gradients. To address the challenges caused by changeable bit widths and temporal lengths, we propose the refined spiking neuron, which can handle different temporal lengths, enable the derivation of gradients for temporal lengths, and suit spike quantization better. In addition, we theoretically formulate the step-size mismatch problem of learnable bit widths, which may incur severe quantization errors to SNN, and accordingly propose the step-size renewal mechanism to alleviate this issue. Experiments on various datasets, including the static CIFAR and ImageNet datasets and the dynamic CIFAR-DVS and DVS-GESTURE datasets, demonstrate that our methods can reduce the overall memory and computation cost while achieving higher accuracy. Particularly, our SEWResNet-34 can achieve a 2.69\% accuracy gain and 4.16$\times$ lower bit budgets over the advanced baseline work on ImageNet. This work is open-sourced at \href{https://github.com/Ikarosy/Towards-Efficient-and-Accurate-Spiking-Neural-Networks-via-Adaptive-Bit-Allocation}{this link}.

cs.NE↗

arXiv2025

Scalable Autoregressive Deep Surrogates for Dendritic Microstructure Dynamics

Microstructural pattern formation, such as dendrite growth, occurs widely in materials and energy systems, significantly influencing material properties and functional performance. While the phase-field method has emerged as a powerful computational tool for modeling microstructure dynamics, its high computational cost limits its integration into practical materials design workflows. Here, we introduce a machine-learning framework using autoregressive deep surrogates trained on short trajectories from quantitative phase-field simulations of alloy solidification in limited spatial domains. Once trained, these surrogates accurately predict dendritic evolution at scalable length and time scales, achieving a speed-up of more than two orders of magnitude. Demonstrations in isothermal growth and in directional solidification of a dilute Al-Cu alloy validate their ability to predict microstructure evolution. Quantitative comparisons with phase-field benchmarks further show excellent agreement in the tip-selection constant, morphological symmetry, and primary spacing evolution.

cond-mat.mtrl-sci↗

arXiv2025

Surface Topological Quantum Criticality II: Conformal manifolds, Isolated fixed points and Entanglement

In this article, we propose the realization of conformal manifolds, which are smooth manifolds of scale-conformal invariant interacting Hamiltonians in two-dimensional quantum many-body systems. Such phenomena can occur in various interacting systems, including topological surfaces or 2D bulks. Building on previous observations, we demonstrate that a conformal manifold can emerge as an exact solution when the number of fermion colors, $N_c$, becomes infinite. We identify distinct exact marginal deformation operators uniquely associated with the conformal manifolds. By considering $N_c$ as finite but large, we show that quantum fluctuations induce a fermion field renormalization that results in mildly infrared relevant or irrelevant renormalization-group (RG) flow within a conformal manifold, producing standard isolated infrared stable Wilson-Fisher fixed points. These can be grouped with ultraviolet stable fixed points into a discrete manifold due to the spontaneous symmetry breaking of an emergent $SO(\mathcal{N})$ dynamical symmetry in the RG flow as $N_c \rightarrow \infty$. Additionally, we find a correlation between the direction of the RG flow within the manifold and an increase in EPR-like entanglement entropy. The infrared-stable Wilson-Fisher fixed points, induced by quantum fluctuations, are linked to theories on the conformal manifold where interaction operators are maximally entangled in flavor space. Our studies provide an effective framework for addressing topological quantum critical points with high-dimensional interaction parameter spaces, potentially housing many fixed points of various stabilities. They also highlight the central role of entangled conformal operators and their entropy in shaping universality classes of surface topological quantum phase transitions. We conclude with open questions and possible future directions.

cond-mat.str-el↗

arXiv2025

PRISM: Progressive Rain removal with Integrated State-space Modeling

Image deraining is an essential vision technique that removes rain streaks and water droplets, enhancing clarity for critical vision tasks like autonomous driving. However, current single-scale models struggle with fine-grained recovery and global consistency. To address this challenge, we propose Progressive Rain removal with Integrated State-space Modeling (PRISM), a progressive three-stage framework: Coarse Extraction Network (CENet), Frequency Fusion Network (SFNet), and Refine Network (RNet). Specifically, CENet and SFNet utilize a novel Hybrid Attention UNet (HA-UNet) for multi-scale feature aggregation by combining channel attention with windowed spatial transformers. Moreover, we propose Hybrid Domain Mamba (HDMamba) for SFNet to jointly model spatial semantics and wavelet domain characteristics. Finally, RNet recovers the fine-grained structures via an original-resolution subnetwork. Our model learns high-frequency rain characteristics while preserving structural details and maintaining global context, leading to improved image quality. Our method achieves competitive results on multiple datasets against recent deraining methods.

cs.CV↗