Searcharxiv⌕ Search

arXiv subjects

Jie Liu

Publications and source records attributed to Jie Liu.

At least 163 records · Page 9Linked to original sources
arXiv2026

Towards Uniformity and Alignment for Multimodal Representation Learning

Multimodal representation learning aims to construct a shared embedding space in which heterogeneous modalities are semantically aligned. Despite strong empirical results, InfoNCE-based objectives introduce inherent conflicts that yield distribution gaps across modalities. In this work, we identify two conflicts in the multimodal regime, both exacerbated as the number of modalities increases: (i) an alignment-uniformity conflict, whereby the repulsion of uniformity undermines pairwise alignment, and (ii) an intra-alignment conflict, where aligning multiple modalities induces competing alignment directions. To address these issues, we propose a principled decoupling of alignment and uniformity for multimodal representations, providing a conflict-free recipe for multimodal learning that simultaneously supports discriminative and generative use cases without task-specific modules. We then provide a theoretical guarantee that our method acts as an efficient proxy for a global Hölder divergence over multiple modality distributions, and thus reduces the distribution gap among modalities. Extensive experiments on retrieval and UnCLIP-style generation demonstrate consistent gains.

cs.LG↗
arXiv2026

Decoherence-free interaction and maximally entangled state generation in giant-atom semi-infinite waveguide systems

Giant atoms are artificial atoms that can couple to a waveguide non-locally. Previous works have shown that two giant atoms in a braided configuration can interact through one-dimensional (1D) infinite and chiral waveguides, with both individual and collective atomic relaxation being fully suppressed. In this paper, however, we show that the decoherence-free interaction (DFI) between two giant atoms can be realized in both braided and nested configurations when the waveguide is semi-infinite. This protected interaction fails to appear in semi-infinite waveguide systems containing two separate giant atoms or two small atoms. We also study the entanglement generation between two giant atoms coupled to a 1D semi-infinite waveguide. The results show that the maximally entangled state is generated in both braided and nested configurations due to the formation of DFI, and in the separate configuration, the maximally achievable entanglement can exceed 0.5. Finally, we generalize the discussion on DFI and entanglement generation to the case involving multiple giant atoms coupled into a semi-infinite waveguide. This study presents a new scheme for realizing DFI and generating maximally entangled states in giant-atom waveguide-QED systems.

quant-ph↗
arXiv2026

Synthetic topological device for advancing elastic energy harvesting

High-efficiency energy harvesting of ultrasonic elastic waves are crucial for powering electric gadgets in many emerging technologies such as wearable devices, wireless sensing, and biomedical implants. Although topological phononic metamaterials have recently been demonstrated as a promising paradigm for confining and guiding elastic waves through robust bound states, achieving ultrahigh-Q topological resonance with enhanced energy conversion efficiency remains a challenge. In this work, we propose a synthetic-dimensional higher-order topological insulator by engineering the flexural bands of elastic metamaterials, featuring highly localized topological hinge states in the bulk bands. This topological hinge mode stems from the nonzero combination of the bulk polarization and the Chern number in the synthetic-dimensional band structure, thus giving rise to a strong elastic-to-electric energy conversion at the corner of the phononic plate. Through numerical simulations and experimental validations, straightforward evidence of the localized modes with robust protection and consequent abilities in activating the light-emitting diodes (LEDs) array have been demonstrated. Our findings open a new avenue for topological-physics-enabled ultrasonic devices and present promising prospects for applications in weak-signal detection and self-powered sensors.

physics.optics↗
arXiv2026

IT-OSE: Exploring Optimal Sample Size for Industrial Data Augmentation

In industrial scenarios, data augmentation is an effective approach to improve model performance. However, its benefits are not unidirectionally beneficial. There is no theoretical research or established estimation for the optimal sample size (OSS) in augmentation, nor is there an established metric to evaluate the accuracy of OSS or its deviation from the ground truth. To address these issues, we propose an information-theoretic optimal sample size estimation (IT-OSE) to provide reliable OSS estimation for industrial data augmentation. An interval coverage and deviation (ICD) score is proposed to evaluate the estimated OSS intuitively. The relationship between OSS and dominant factors is theoretically analyzed and formulated, thereby enhancing the interpretability. Experiments show that, compared to empirical estimation, the IT-OSE increases accuracy in classification tasks across baseline models by an average of 4.38%, and reduces MAPE in regression tasks across baseline models by an average of 18.80%. The improvements in downstream model performance are more stable. ICDdev in the ICD score is also reduced by an average of 49.30%. The determinism of OSS is enhanced. Compared to exhaustive search, the IT-OSE achieves the same OSS while reducing computational and data costs by an average of 83.97% and 93.46%. Furthermore, practicality experiments demonstrate that the IT-OSE exhibits generality across representative sensor-based industrial scenarios.

cs.LG↗
arXiv2026

Quantized-Tinyllava: a new multimodal foundation model enables efficient split learning

Multimodal foundation models are increasingly trained on sensitive data across domains such as finance, biomedicine, and personal identifiers. However, this distributed setup raises serious privacy concerns due to the need for cross-partition data sharing. Split learning addresses these concerns by enabling collaborative model training without raw data exchange between partitions, yet it introduces a significant challenge: transmitting high-dimensional intermediate feature representations between partitions leads to substantial communication costs. To address this challenge, we propose Quantized-TinyLLaVA, a multimodal foundation model with an integrated communication-efficient split learning framework. Our approach adopts a compression module that quantizes intermediate feature into discrete representations before transmission, substantially reducing communication overhead. Besides, we derive a principled quantization strategy grounded in entropy coding theory to determine the optimal number of discrete representation levels. We deploy our framework in a two-partition setting, with one partition operating as the client and the other as the server, to realistically simulate distributed training. Under this setup, Quantized-TinyLLaVA achieves an approximate \textbf{87.5\%} reduction in communication overhead with 2-bit quantization, while maintaining performance of the original 16-bit model across five benchmark datasets. Furthermore, our compressed representations exhibit enhanced resilience against feature inversion attacks, validating the privacy of transmission. The code is available at https://github.com/anonymous-1742/Quantized-TinyLLaVA.

cs.LG↗
arXiv2026

Riesz transforms for the distinguished Laplacian on Damek-Ricci spaces and operator-valued multivariate spectral multipliers

Let $Δ= \nabla^* \nabla$ be the distinguished Laplacian on a Damek-Ricci space. We prove the $L^{p}$-boundedness of the vector of first-order Riesz transforms $\nabla Δ^{-1/2}$ in the full range $p\in(1,\infty)$. The most demanding part of the proof is establishing the boundedness for $p \in (2,\infty)$; this is obtained as a consequence of an operator-valued spectral multiplier theorem for the joint functional calculus of a commuting system of self-adjoint operators, which we prove here and may be of independent interest.

math.FA↗
arXiv2026

GLAD: Generative Language-Assisted Visual Tracking for Low-Semantic Templates

Vision-language tracking has gained increasing attention in many scenarios. This task simultaneously deals with visual and linguistic information to localize objects in videos. Despite its growing utility, the development of vision-language tracking methods remains in its early stage. Current vision-language trackers usually employ Transformer architectures for interactive integration of template, search, and text features. However, persistent challenges about low-semantic images including prevalent image blurriness, low resolution and so on, may compromise model performance through degraded cross-modal understanding. To solve this problem, language assistance is usually used to deal with the obstacles posed by low-semantic images. However, due to the existing gap between current textual and visual features, direct concatenation and fusion of these features may have limited effectiveness. To address these challenges, we introduce a pioneering Generative Language-AssisteD tracking model, GLAD, which utilizes diffusion models for the generative multi-modal fusion of text description and template image to bolster compatibility between language and image and enhance template image semantic information. Our approach demonstrates notable improvements over the existing fusion paradigms. Blurry and semantically ambiguous template images can be restored to improve multi-modal features in the generative fusion paradigm. Experiments show that our method establishes a new state-of-the-art on multiple benchmarks and achieves an impressive inference speed. The code and models will be released at: https://github.com/Confetti-lxy/GLAD

cs.CV↗
arXiv2026

HSI-VAR: Rethinking Hyperspectral Restoration through Spatial-Spectral Visual Autoregression

Hyperspectral images (HSIs) capture richer spatial-spectral information beyond RGB, yet real-world HSIs often suffer from a composite mix of degradations, such as noise, blur, and missing bands. Existing generative approaches for HSI restoration like diffusion models require hundreds of iterative steps, making them computationally impractical for high-dimensional HSIs. While regression models tend to produce oversmoothed results, failing to preserve critical structural details. We break this impasse by introducing HSI-VAR, rethinking HSI restoration as an autoregressive generation problem, where spectral and spatial dependencies can be progressively modeled rather than globally reconstructed. HSI-VAR incorporates three key innovations: (1) Latent-condition alignment, which couples semantic consistency between latent priors and conditional embeddings for precise reconstruction; (2) Degradation-aware guidance, which uniquely encodes mixed degradations as linear combinations in the embedding space for automatic control, remarkably achieving a nearly $50\%$ reduction in computational cost at inference; (3) A spatial-spectral adaptation module that refines details across both domains in the decoding phase. Extensive experiments on nine all-in-one HSI restoration benchmarks confirm HSI-VAR's state-of-the-art performance, achieving a 3.77 dB PSNR improvement on \textbf{\textit{ICVL}} and offering superior structure preservation with an inference speed-up of up to $95.5 \times$ compared with diffusion-based methods, making it a highly practical solution for real-world HSI restoration.

cs.CV↗
arXiv2026

Open-Vocabulary Functional 3D Human-Scene Interaction Generation

Generating 3D humans that functionally interact with 3D scenes remains an open problem with applications in embodied AI, robotics, and interactive content creation. The key challenge involves reasoning about both the semantics of functional elements in 3D scenes and the 3D human poses required to achieve functionality-aware interaction. Unfortunately, existing methods typically lack explicit reasoning over object functionality and the corresponding human-scene contact, resulting in implausible or functionally incorrect interactions. In this work, we propose FunHSI, a training-free, functionality-driven framework that enables functionally correct human-scene interactions from open-vocabulary task prompts. Given a task prompt, FunHSI performs functionality-aware contact reasoning to identify functional scene elements, reconstruct their 3D geometry, and model high-level interactions via a contact graph. We then leverage vision-language models to synthesize a human performing the task in the image and estimate proposed 3D body and hand poses. Finally, the proposed 3D body configuration is refined via stage-wise optimization to ensure physical plausibility and functional correctness. In contrast to existing methods, FunHSI not only synthesizes more plausible general 3D interactions, such as "sitting on a sofa'', while supporting fine-grained functional human-scene interactions, e.g., "increasing the room temperature''. Extensive experiments demonstrate that FunHSI consistently generates functionally correct and physically plausible human-scene interactions across diverse indoor and outdoor scenes.

cs.CV↗
arXiv2026

LLM4Fluid: Large Language Models as Generalizable Neural Solvers for Fluid Dynamics

Deep learning has emerged as a promising paradigm for spatio-temporal modeling of fluid dynamics. However, existing approaches often suffer from limited generalization to unseen flow conditions and typically require retraining when applied to new scenarios. In this paper, we present LLM4Fluid, a spatio-temporal prediction framework that leverages Large Language Models (LLMs) as generalizable neural solvers for fluid dynamics. The framework first compresses high-dimensional flow fields into a compact latent space via reduced-order modeling enhanced with a physics-informed disentanglement mechanism, effectively mitigating spatial feature entanglement while preserving essential flow structures. A pretrained LLM then serves as a temporal processor, autoregressively predicting the dynamics of physical sequences with time series prompts. To bridge the modality gap between prompts and physical sequences, which can otherwise degrade prediction accuracy, we propose a dedicated modality alignment strategy that resolves representational mismatch and stabilizes long-term prediction. Extensive experiments across diverse flow scenarios demonstrate that LLM4Fluid functions as a robust and generalizable neural solver without retraining, achieving state-of-the-art accuracy while exhibiting powerful zero-shot and in-context learning capabilities. Code and datasets are publicly available at https://github.com/qisongxiao/LLM4Fluid.

cs.LG↗
arXiv2026

Advancing Open-source World Models

We present LingBot-World, an open-sourced world simulator stemming from video generation. Positioned as a top-tier world model, LingBot-World offers the following features. (1) It maintains high fidelity and robust dynamics in a broad spectrum of environments, including realism, scientific contexts, cartoon styles, and beyond. (2) It enables a minute-level horizon while preserving contextual consistency over time, which is also known as "long-term memory". (3) It supports real-time interactivity, achieving a latency of under 1 second when producing 16 frames per second. We provide public access to the code and model in an effort to narrow the divide between open-source and closed-source technologies. We believe our release will empower the community with practical applications across areas like content creation, gaming, and robot learning.

cs.CV↗
arXiv2026

The affine closure of cotangent bundles of horospherical spaces

For a smooth quasi-affine variety $X$, the affine closure $\overline{T^*X} := \text{Spec}(\mathbb{K}[T^*X])$ contains $T^*X$ as an open subset, and its smooth locus carries a symplectic structure. A natural question is whether $\overline{T^*X}$ itself is a symplectic variety. A notable example is the conjecture of Ginzburg and Kazhdan, which predicts that $\overline{T^*(G/U)}$ is symplectic for a maximal unipotent subgroup $U$ in a reductive linear algebraic group $G$. This conjecture was recently proved by Gannon using representation-theoretic methods. In this paper, we provide a new geometric approach to this conjecture. Our method allows us to prove a more general result: $\overline{T^*(G/H)}$ is symplectic for any horospherical subgroup $H$ in $G$ such that $G/H$ is quasi-affine. In particular, this implies that the affine closure $\overline{T^*(G/[P,P])}$ is a symplectic variety for any parabolic subgroup $P$ in $G$.

math.AG↗
arXiv2026

Hamiltonian reductions as affine closures of cotangent bundles

Let $Y$ be an irreducible non-singular affine $G$-variety with a $2$-large action. We show that the Hamiltonian reduction $T^*Y/\!\!/\!\!/G$ is a symplectic variety with terminal singularities, isomorphic to the affine closure of $T^*Z_{\text{reg}}$ where $Z:=Y/\!/G$. Furthermore, we provide sufficient conditions for the non-existence of a symplectic resolution for such varieties. These results yield three main applications: (i) providing a short proof of G. Schwarz's theorem on the graded surjectivity of the push-forward map $\mathcal{D}(Y)^G \to \mathcal{D}(Z)$; (ii) establishing the surjectivity of the symbol map on $Z$; and (iii) confirming the non-linear analog of a conjecture of Kaledin--Lehn--Sorger for $2$-large actions.

math.AG↗
arXiv2026

Skywork UniPic 2.0: Building Kontext Model with Online RL for Unified Multimodal Model

Recent advances in multimodal models have demonstrated impressive capabilities in unified image generation and editing. However, many prominent open-source models prioritize scaling model parameters over optimizing training strategies, limiting their efficiency and performance. In this work, we present UniPic2-SD3.5M-Kontext, a 2B-parameter DiT model based on SD3.5-Medium, which achieves state-of-the-art image generation and editing while extending seamlessly into a unified multimodal framework. Our approach begins with architectural modifications to SD3.5-Medium and large-scale pre-training on high-quality data, enabling joint text-to-image generation and editing capabilities. To enhance instruction following and editing consistency, we propose a novel Progressive Dual-Task Reinforcement strategy (PDTR), which effectively strengthens both tasks in a staged manner. We empirically validate that the reinforcement phases for different tasks are mutually beneficial and do not induce negative interference. After pre-training and reinforcement strategies, UniPic2-SD3.5M-Kontext demonstrates stronger image generation and editing capabilities than models with significantly larger generation parameters-including BAGEL (7B) and Flux-Kontext (12B). Furthermore, following the MetaQuery, we connect the UniPic2-SD3.5M-Kontext and Qwen2.5-VL-7B via a connector and perform joint training to launch a unified multimodal model UniPic2-Metaquery. UniPic2-Metaquery integrates understanding, generation, and editing, achieving top-tier performance across diverse tasks with a simple and scalable training paradigm. This consistently validates the effectiveness and generalizability of our proposed training paradigm, which we formalize as Skywork UniPic 2.0.

cs.CV↗
arXiv2026

VoiceSculptor: Your Voice, Designed By You

Despite rapid progress in text-to-speech (TTS), open-source systems still lack truly instruction-following, fine-grained control over core speech attributes (e.g., pitch, speaking rate, age, emotion, and style). We present VoiceSculptor, an open-source unified system that bridges this gap by integrating instruction-based voice design and high-fidelity voice cloning in a single framework. It generates controllable speaker timbre directly from natural-language descriptions, supports iterative refinement via Retrieval-Augmented Generation (RAG), and provides attribute-level edits across multiple dimensions. The designed voice is then rendered into a prompt waveform and fed into a cloning model to enable high-fidelity timbre transfer for downstream speech synthesis. VoiceSculptor achieves open-source state-of-the-art (SOTA) on InstructTTSEval-Zh, and is fully open-sourced, including code and pretrained models, to advance reproducible instruction-controlled TTS research.

eess.AS↗
arXiv2026

Current-driven nonlinear skyrmion dynamics in altermagnets

The center of mass and helicity are two dynamic degrees of freedom of skyrmions. In this work, we study the current-driven skyrmion motion in frustrated altermagnets. Contrary to conventional wisdom, we find that the skyrmion helicity is not locked with the skyrmion Hall angle, but unidirectionally rotates with a global angular velocity proportional to the square of the current density. In addition, we find that the helicity rotation velocity is highly anisotropic, depending on the direction of current flows. We also observe helicity oscillation in the terahertz regimes, where the nonlinear mixing between the fast and slow modes generates a magnon frequency comb. Full atomistic spin dynamics simulations verify our theoretical predictions. Our results establish frustrated altermagnets as a promising platform for skyrmionics, THz technology, and frequency comb.

cond-mat.mtrl-sci↗
arXiv2026

Paired Image Generation with Diffusion-Guided Diffusion Models

The segmentation of mass lesions in digital breast tomosynthesis (DBT) images is very significant for the early screening of breast cancer. However, the high-density breast tissue often leads to high concealment of the mass lesions, which makes manual annotation difficult and time-consuming. As a result, there is a lack of annotated data for model training. Diffusion models are commonly used for data augmentation, but the existing methods face two challenges. First, due to the high concealment of lesions, it is difficult for the model to learn the features of the lesion area. This leads to the low generation quality of the lesion areas, thus limiting the quality of the generated images. Second, existing methods can only generate images and cannot generate corresponding annotations, which restricts the usability of the generated images in supervised training. In this work, we propose a paired image generation method. The method does not require external conditions and can achieve the generation of paired images by training an extra diffusion guider for the conditional diffusion model. During the experimental phase, we generated paired DBT slices and mass lesion masks. Then, we incorporated them into the supervised training process of the mass lesion segmentation task. The experimental results show that our method can improve the generation quality without external conditions. Moreover, it contributes to alleviating the shortage of annotated data, thus enhancing the performance of downstream tasks. The source code is available at https://github.com/zhanghx1320/PIG.

cs.CV↗
arXiv2026

Sel-assembled Rhodium Nanoantennas for Single-Protein UV SERS

Surface-enhanced Raman scattering (SERS) provides critical insights into analyte structure, dynamic processes, and intermolecular interactions at the single-molecule level. By exploiting the hotspot formation in the vicinity of plasmonic structures, SERS constitutes an established tool for fundamental biological research, particularly for early-stage disease diagnostics. In this context, the DNA Origami technique, with its high addressability, enables both the assembly of plasmonic nanostructures with nanometric accuracy, and the deterministic placement of a single analyte molecule precisely at the generated hotspot within them. To date, most DNA Origami based nanoantennas rely on gold or silver nanoparticles (NPs), whose plasmonic resonances are confined to the visible spectrum, severely limiting their use in other spectral ranges. To extend the operating range, we have recently established a robust strategy for self-assembling programmable ultraviolet (UV)-plasmonic dimer antennas using rhodium nanocubes. Herein, we leverage this tailored architecture to systematically investigate its performance for single-molecule UV-SERS. We demonstrated how biofabricated Rh-dimers can be used to detect the characteristic SERS signal of a single streptavidin molecule linked at the dimer s gap. Our results are validated through polarization dependent measurements that yield the expected signal modulation depending on the the dimer orientation only for the DNA origami with a protein at the hotspot. This work establishes a highly sensitive and polarization-tunable UV-SERS platform, laying a solid foundation for label-free optical investigation and bio-spectroscopy of individual biomolecules in the UV spectral range.

physics.app-ph↗