arXiv subjects

Lei Wang

Publications and source records attributed to Lei Wang.

At least 361 records · Page 20Linked to original sources

arXiv2025

Prototype-Guided Curriculum Learning for Zero-Shot Learning

In Zero-Shot Learning (ZSL), embedding-based methods enable knowledge transfer from seen to unseen classes by learning a visual-semantic mapping from seen-class images to class-level semantic prototypes (e.g., attributes). However, these semantic prototypes are manually defined and may introduce noisy supervision for two main reasons: (i) instance-level mismatch: variations in perspective, occlusion, and annotation bias will cause discrepancies between individual sample and the class-level semantic prototypes; and (ii) class-level imprecision: the manually defined semantic prototypes may not accurately reflect the true semantics of the class. Consequently, the visual-semantic mapping will be misled, reducing the effectiveness of knowledge transfer to unseen classes. In this work, we propose a prototype-guided curriculum learning framework (dubbed as CLZSL), which mitigates instance-level mismatches through a Prototype-Guided Curriculum Learning (PCL) module and addresses class-level imprecision via a Prototype Update (PUP) module. Specifically, the PCL module prioritizes samples with high cosine similarity between their visual mappings and the class-level semantic prototypes, and progressively advances to less-aligned samples, thereby reducing the interference of instance-level mismatches to achieve accurate visual-semantic mapping. Besides, the PUP module dynamically updates the class-level semantic prototypes by leveraging the visual mappings learned from instances, thereby reducing class-level imprecision and further improving the visual-semantic mapping. Experiments were conducted on standard benchmark datasets-AWA2, SUN, and CUB-to verify the effectiveness of our method.

cs.CV↗

arXiv2025

Dynamic Hologram Generation with Automatic Differentiation

We designed an automatic differentiation-based strategy to generate optical trap arrays that change smoothly in time. Instead of repeatedly regenerating the holograms for each time step, we derive the differential form of the phase dynamics that enables the continuous evolution of the trap coordinates. This differential form is derived from the implicit differentiation of the fixed point of the Gerchberg-Saxton algorithm, which is computationally efficient. We carried out numerical and laboratory experiments to demonstrate its effectiveness in improving the phase continuity and reducing the computational burden compared to the traditional pure interpolation techniques. By combining the method with the spatial light modulator, the method is promising for the dynamic manipulation of particles in real experiments.

physics.optics↗

arXiv2025

HSA-Net: Hierarchical and Structure-Aware Framework for Efficient and Scalable Molecular Language Modeling

Molecular representation learning, a cornerstone for downstream tasks like molecular captioning and molecular property prediction, heavily relies on Graph Neural Networks (GNN). However, GNN suffers from the over-smoothing problem, where node-level features collapse in deep GNN layers. While existing feature projection methods with cross-attention have been introduced to mitigate this issue, they still perform poorly in deep features. This motivated our exploration of using Mamba as an alternative projector for its ability to handle complex sequences. However, we observe that while Mamba excels at preserving global topological information from deep layers, it neglects fine-grained details in shallow layers. The capabilities of Mamba and cross-attention exhibit a global-local trade-off. To resolve this critical global-local trade-off, we propose Hierarchical and Structure-Aware Network (HSA-Net), a novel framework with two modules that enables a hierarchical feature projection and fusion. Firstly, a Hierarchical Adaptive Projector (HAP) module is introduced to process features from different graph layers. It learns to dynamically switch between a cross-attention projector for shallow layers and a structure-aware Graph-Mamba projector for deep layers, producing high-quality, multi-level features. Secondly, to adaptively merge these multi-level features, we design a Source-Aware Fusion (SAF) module, which flexibly selects fusion experts based on the characteristics of the aggregation features, ensuring a precise and effective final representation fusion. Extensive experiments demonstrate that our HSA-Net framework quantitatively and qualitatively outperforms current state-of-the-art (SOTA) methods.

cs.LG↗

arXiv2025

BOASF: A Unified Framework for Speeding up Automatic Machine Learning via Adaptive Successive Filtering

Machine learning has been making great success in many application areas. However, for the non-expert practitioners, it is always very challenging to address a machine learning task successfully and efficiently. Finding the optimal machine learning model or the hyperparameter combination set from a large number of possible alternatives usually requires considerable expert knowledge and experience. To tackle this problem, we propose a combined Bayesian Optimization and Adaptive Successive Filtering algorithm (BOASF) under a unified multi-armed bandit framework to automate the model selection or the hyperparameter optimization. Specifically, BOASF consists of multiple evaluation rounds in each of which we select promising configurations for each arm using the Bayesian optimization. Then, ASF can early discard the poor-performed arms adaptively using a Gaussian UCB-based probabilistic model. Furthermore, a Softmax model is employed to adaptively allocate available resources for each promising arm that advances to the next round. The arm with a higher probability of advancing will be allocated more resources. Experimental results show that BOASF is effective for speeding up the model selection and hyperparameter optimization processes while achieving robust and better prediction performance than the existing state-of-the-art automatic machine learning methods. Moreover, BOASF achieves better anytime performance under various time budgets.

cs.LG↗

arXiv2025

Explainable Recommendation with Simulated Human Feedback

Recent advancements in explainable recommendation have greatly bolstered user experience by elucidating the decision-making rationale. However, the existing methods actually fail to provide effective feedback signals for potentially better or worse generated explanations due to their reliance on traditional supervised learning paradigms in sparse interaction data. To address these issues, we propose a novel human-like feedback-driven optimization framework. This framework employs a dynamic interactive optimization mechanism for achieving human-centered explainable requirements without incurring high labor costs. Specifically, we propose to utilize large language models (LLMs) as human simulators to predict human-like feedback for guiding the learning process. To enable the LLMs to deeply understand the task essence and meet user's diverse personalized requirements, we introduce a human-induced customized reward scoring method, which helps stimulate the language understanding and logical reasoning capabilities of LLMs. Furthermore, considering the potential conflicts between different perspectives of explanation quality, we introduce a principled Pareto optimization that transforms the multi-perspective quality enhancement task into a multi-objective optimization problem for improving explanation performance. At last, to achieve efficient model training, we design an off-policy optimization pipeline. By incorporating a replay buffer and addressing the data distribution biases, we can effectively improve data utilization and enhance model generality. Extensive experiments on four datasets demonstrate the superiority of our approach.

cs.IR↗

arXiv2025

Anomalous charge density wave in altermagnetism

Exploring the intricate interplay between magnetism and charge density waves has long been a fundamental pursuit at the forefront of condensed matter research. In this letter, based on symmetry analysis and first-principles calculations, we propose for the first time that anomalous charge density wave can be realized in two-dimensional altermagnetic WO. The anomalous charge density wave is characterized by three key features: (i) Unlike conventional charge density wave, whose stabilization is driven by the opening of a gap near the Fermi level, the anomalous charge density wave is stabilized by the occupied states with energies shifting lower far away from the Fermi level; (ii) the anomalous charge density wave increases the density of states near the Fermi level and then enhances-rather than diminishes-the metallicity of materials; (iii) altermagnetism plays a crucial role in stabilizing anomalous charge density wave. Thus, our work offers a pathway for exploring both the realization and the underlying mechanisms of anomalous charge density waves in magnetic systems.

cond-mat.mtrl-sci↗

arXiv2025

S-RRG-Bench: Structured Radiology Report Generation with Fine-Grained Evaluation Framework

Radiology report generation (RRG) for diagnostic images, such as chest X-rays, plays a pivotal role in both clinical practice and AI. Traditional free-text reports suffer from redundancy and inconsistent language, complicating the extraction of critical clinical details. Structured radiology report generation (S-RRG) offers a promising solution by organizing information into standardized, concise formats. However, existing approaches often rely on classification or visual question answering (VQA) pipelines that require predefined label sets and produce only fragmented outputs. Template-based approaches, which generate reports by replacing keywords within fixed sentence patterns, further compromise expressiveness and often omit clinically important details. In this work, we present a novel approach to S-RRG that includes dataset construction, model training, and the introduction of a new evaluation framework. We first create a robust chest X-ray dataset (MIMIC-STRUC) that includes disease names, severity levels, probabilities, and anatomical locations, ensuring that the dataset is both clinically relevant and well-structured. We train an LLM-based model to generate standardized, high-quality reports. To assess the generated reports, we propose a specialized evaluation metric (S-Score) that not only measures disease prediction accuracy but also evaluates the precision of disease-specific details, thus offering a clinically meaningful metric for report quality that focuses on elements critical to clinical decision-making and demonstrates a stronger alignment with human assessments. Our approach highlights the effectiveness of structured reports and the importance of a tailored evaluation metric for S-RRG, providing a more clinically relevant measure of report quality.

cs.CV↗

arXiv2025

CX-Mind: A Pioneering Multimodal Large Language Model for Interleaved Reasoning in Chest X-ray via Curriculum-Guided Reinforcement Learning

Chest X-ray (CXR) imaging is one of the most widely used diagnostic modalities in clinical practice, encompassing a broad spectrum of diagnostic tasks. Recent advancements have seen the extensive application of reasoning-based multimodal large language models (MLLMs) in medical imaging to enhance diagnostic efficiency and interpretability. However, existing multimodal models predominantly rely on "one-time" diagnostic approaches, lacking verifiable supervision of the reasoning process. This leads to challenges in multi-task CXR diagnosis, including lengthy reasoning, sparse rewards, and frequent hallucinations. To address these issues, we propose CX-Mind, the first generative model to achieve interleaved "think-answer" reasoning for CXR tasks, driven by curriculum-based reinforcement learning and verifiable process rewards (CuRL-VPR). Specifically, we constructed an instruction-tuning dataset, CX-Set, comprising 708,473 images and 2,619,148 samples, and generated 42,828 high-quality interleaved reasoning data points supervised by clinical reports. Optimization was conducted in two stages under the Group Relative Policy Optimization framework: initially stabilizing basic reasoning with closed-domain tasks, followed by transfer to open-domain diagnostics, incorporating rule-based conditional process rewards to bypass the need for pretrained reward models. Extensive experimental results demonstrate that CX-Mind significantly outperforms existing medical and general-domain MLLMs in visual understanding, text generation, and spatiotemporal alignment, achieving an average performance improvement of 25.1% over comparable CXR-specific models. On real-world clinical dataset (Rui-CXR), CX-Mind achieves a mean recall@1 across 14 diseases that substantially surpasses the second-best results, with multi-center expert evaluations further confirming its clinical utility across multiple dimensions.

cs.LG↗

arXiv2025

Talking-to-Build: How LLM-Assisted Interface Shapes Player Performance and Experience in Minecraft

With large language models (LLMs) on the rise, in-game interactions are shifting from rigid commands to natural conversations. However, the impacts of LLMs on player performance and game experience remain underexplored. This work explores LLM's role as a co-builder during gameplay, examining its impact on task performance, usability, and player experience. Using Minecraft as a sandbox, we present an LLM-assisted interface that engages players through natural language, aiming to facilitate creativity and simplify complex gaming commands. We conducted a mixed-methods study with 30 participants, comparing LLM-assisted and command-based interfaces across simple and complex game tasks. Quantitative and qualitative analyses reveal that the LLM-assisted interface significantly improves player performance, engagement, and overall game experience. Additionally, task complexity has a notable effect on player performance and experience across both interfaces. Our findings highlight the potential of LLM-assisted interfaces to revolutionize virtual experiences, emphasizing the importance of balancing intuitiveness with predictability, transparency, and user agency in AI-driven, multimodal gaming environments.

cs.HC↗

arXiv2025

Enhancing interfacial thermal conductance in Si/Diamond heterostructures by phonon bridge

This study investigates the mechanism of enhancing interfacial thermal transport performance in Silicon/Diamond (Si/Diamond) heterostructures using the phonon bridge. A heat transfer model for three-layer heterostructures is developed by combining First-principles calculations with the Monte Carlo method. The temperature distribution, spectral heat conductance, and interfacial thermal conductance are compared for Si/Diamond heterostructures with and without a silicon carbide (SiC) interlayer. The results show that the SiC interlayer effectively bridges low-frequency phonons in Si with mid-to-high-frequency phonons in Diamond, which forms a specific phonon bridge, significantly improving interfacial phonon transport. The influence of SiC interlayer thickness is further studied, revealing a size-dependent phonon bridge enhancement. For thin interlayers, intensified phonon boundary scattering weakens the bridging effect. Conversely, excessively thick interlayers increase the bulk thermal resistance, reducing overall interfacial thermal conductance. Thus, an optimal interlayer thickness exists, identified as 40 nm for SiC. Thirteen candidate interlayer materials, including SiC, AlN, α-Si3N4, \b{eta}-Si3N4, and AlxGa1-xN (x ranges from 0.1 to 0.9), are compared at various thicknesses. SiC emerges as the most effective interlayer material, increasing interfacial thermal conductance by 46.6% compared to the bilayer heterostructure. AlN ranks second, improving thermal conductance by 21.9%. These findings provide essential insights into the phonon bridge mechanism at heterogeneous interface thermal transport and offer valuable theoretical guidance for designing heterostructures with enhanced thermal transport performance.

cond-mat.mtrl-sci↗

arXiv2025

LUT Tensor Core: A Software-Hardware Co-Design for LUT-Based Low-Bit LLM Inference

Large Language Model (LLM) inference becomes resource-intensive, prompting a shift toward low-bit model weights to reduce the memory footprint and improve efficiency. Such low-bit LLMs necessitate the mixed-precision matrix multiplication (mpGEMM), an important yet underexplored operation involving the multiplication of lower-precision weights with higher-precision activations. Off-the-shelf hardware does not support this operation natively, leading to indirect, thus inefficient, dequantization-based implementations. In this paper, we study the lookup table (LUT)-based approach for mpGEMM and find that a conventional LUT implementation fails to achieve the promised gains. To unlock the full potential of LUT-based mpGEMM, we propose LUT Tensor Core, a software-hardware co-design for low-bit LLM inference. LUT Tensor Core differentiates itself from conventional LUT designs through: 1) software-based optimizations to minimize table precompute overhead and weight reinterpretation to reduce table storage; 2) a LUT-based Tensor Core hardware design with an elongated tiling shape to maximize table reuse and a bit-serial design to support diverse precision combinations in mpGEMM; 3) a new instruction set and compilation optimizations for LUT-based mpGEMM. LUT Tensor Core significantly outperforms existing pure software LUT implementations and achieves a 1.44$\times$ improvement in compute density and energy efficiency compared to previous state-of-the-art LUT-based accelerators.

cs.AR↗

arXiv2025

OpenHuman4D: Open-Vocabulary 4D Human Parsing

Understanding dynamic 3D human representation has become increasingly critical in virtual and extended reality applications. However, existing human part segmentation methods are constrained by reliance on closed-set datasets and prolonged inference times, which significantly restrict their applicability. In this paper, we introduce the first 4D human parsing framework that simultaneously addresses these challenges by reducing the inference time and introducing open-vocabulary capabilities. Building upon state-of-the-art open-vocabulary 3D human parsing techniques, our approach extends the support to 4D human-centric video with three key innovations: 1) We adopt mask-based video object tracking to efficiently establish spatial and temporal correspondences, avoiding the necessity of segmenting all frames. 2) A novel Mask Validation module is designed to manage new target identification and mitigate tracking failures. 3) We propose a 4D Mask Fusion module, integrating memory-conditioned attention and logits equalization for robust embedding fusion. Extensive experiments demonstrate the effectiveness and flexibility of the proposed method on 4D human-centric parsing tasks, achieving up to 93.3% acceleration compared to the previous state-of-the-art method, which was limited to parsing fixed classes.

cs.CV↗

arXiv2025

Modeling enzyme temperature stability from sequence segment perspective

Developing enzymes with desired thermal properties is crucial for a wide range of industrial and research applications, and determining temperature stability is an essential step in this process. Experimental determination of thermal parameters is labor-intensive, time-consuming, and costly. Moreover, existing computational approaches are often hindered by limited data availability and imbalanced distributions. To address these challenges, we introduce a curated temperature stability dataset designed for model development and benchmarking in enzyme thermal modeling. Leveraging this dataset, we present the \textit{Segment Transformer}, a novel deep learning framework that enables efficient and accurate prediction of enzyme temperature stability. The model achieves state-of-the-art performance with an RMSE of 24.03, MAE of 18.09, and Pearson and Spearman correlations of 0.33, respectively. These results highlight the effectiveness of incorporating segment-level representations, grounded in the biological observation that different regions of a protein sequence contribute unequally to thermal behavior. As a proof of concept, we applied the Segment Transformer to guide the engineering of a cutinase enzyme. Experimental validation demonstrated a 1.64-fold improvement in relative activity following heat treatment, achieved through only 17 mutations and without compromising catalytic function.

cs.LG↗

arXiv2025

Optics design of the Super Tau-Charm Facility collider rings

The Super Tau-Charm Facility (STCF), China's next-generation electron-positron collider, targets an unprecedented luminosity exceeding 5x10^34 cm^-2 s^-1 at a center-of-mass energy of 4 GeV. The implementation of a submillimeter vertical beta function at interaction point (< 1 mm) and crab-waist collision scheme in this low-energy regime introduces critical challenges through severe nonlinear effects that constrain dynamic aperture and degrade Touschek lifetime. To address these constraints, we propose a novel quasi-two-fold symmetric lattice design integrating several synergistic features: Linear optics optimization minimizing the H-invariant around the ring to maximize local momentum acceptance (LMA); Up to third-order of local chromaticity correction in the interaction region combined with second-order achromatic arc optics, enhancing off-momentum beam dynamics; Configured FODO arc structure with interleaved sextupole groups satisfying -I transformation, suppressing third-order geometric aberrations while optimizing Montague function distributions; Advanced final focus system integrating chromatic sextupoles, crab sextupoles, and strategically positioned octupoles to counteract final quadrupole fringe fields. Furthermore, we develop a multi-objective genetic algorithm using the in-house toolkit PAMKIT to simultaneously optimize 46 sextupole families, maximizing both dynamic aperture and momentum bandwidth. Optics performance is evaluated under error conditions with appropriate corrections, ensuring robust beam dynamics.

physics.acc-ph↗

arXiv2025

Multimodal Fine-grained Reasoning for Post Quality Evaluation

Accurately assessing post quality requires complex relational reasoning to capture nuanced topic-post relationships. However, existing studies face three major limitations: (1) treating the task as unimodal categorization, which fails to leverage multimodal cues and fine-grained quality distinctions; (2) introducing noise during deep multimodal fusion, leading to misleading signals; and (3) lacking the ability to capture complex semantic relationships like relevance and comprehensiveness. To address these issues, we propose the Multimodal Fine-grained Topic-post Relational Reasoning (MFTRR) framework, which mimics human cognitive processes. MFTRR reframes post-quality assessment as a ranking task and incorporates multimodal data to better capture quality variations. It consists of two key modules: (1) the Local-Global Semantic Correlation Reasoning Module, which models fine-grained semantic interactions between posts and topics at both local and global levels, enhanced by a maximum information fusion mechanism to suppress noise; and (2) the Multi-Level Evidential Relational Reasoning Module, which explores macro- and micro-level relational cues to strengthen evidence-based reasoning. We evaluate MFTRR on three newly constructed multimodal topic-post datasets and the public Lazada-Home dataset. Experimental results demonstrate that MFTRR significantly outperforms state-of-the-art baselines, achieving up to 9.52% NDCG@3 improvement over the best unimodal method on the Art History dataset.

cs.LG↗

arXiv2025

Distributed Optimization by Network Flows with Spatio-Temporal Compression

Several data compressors have been proposed in distributed optimization frameworks of network systems to reduce communication overhead in large-scale applications. In this paper, we demonstrate that effective information compression may occur over time or space during sequences of node communications in distributed algorithms, leading to the concept of spatio-temporal compressors. This abstraction classifies existing compressors and inspires new compressors as spatio-temporal compressors, with their effectiveness described by constructive stability criteria from nonlinear system theory. Subsequently, we incorporate these spatio-temporal compressors directly into standard continuous-time consensus flows and distributed primal-dual flows, establishing conditions ensuring exponential convergence. Additionally, we introduce a novel observer-based distributed primal-dual continuous flow integrated with spatio-temporal compressors, which provides broader convergence conditions. These continuous flows achieve exponential convergence to the global optimum when the objective function is strongly convex and can be discretized using Euler approximations. Finally, numerical simulations illustrate the versatility of the proposed spatio-temporal compressors and verify the convergence of

eess.SY↗

arXiv2025

Critical angles and one-dimensional moiré physics in twisted rectangular lattices

Engineering moiré superlattices in van der Waals heterostructures provides fundamental control over emergent electronic, structural, and optical properties allowing to affect topological and correlated phenomena. This control is achieved through imposed periodic modulation of potentials and targeted modifications of symmetries. For twisted bilayers of van der Waals materials with rectangular lattices, such as PdSe2, this work shows that one-dimensional (1D) moiré patterns emerge universally. This emergence is driven by a series of critical twist angles (CAs). We investigate the geometric origins of these unique 1D moiré patterns and develop a universal mathematical framework to predict the CAs in twisted rectangular lattices. Through a density functional theory (DFT) description of the electronic properties of twisted bilayer PdSe2, we further reveal directionally localized flat band structures, localized charge densities and strong spin-orbit coupling along the dispersive direction which points to the emergence of an effectively 1D strongly spin-orbit coupled electronic systems. This establishes twisted rectangular systems as a unique platform for engineering low-symmetry moiré patterns, low-dimensional strongly correlated and topological physics, and spatially selective quantum phases beyond the isotropic paradigms of hexagonal moiré materials.

cond-mat.mes-hall↗

arXiv2025

An AI-powered Technology Stack for Solving Many-Electron Field Theory

Quantum field theory (QFT) for interacting many-electron systems is fundamental to condensed matter physics, yet achieving accurate solutions confronts computational challenges in managing the combinatorial complexity of Feynman diagrams, implementing systematic renormalization, and evaluating high-dimensional integrals. We present a unifying framework that integrates QFT computational workflows with an AI-powered technology stack. A cornerstone of this framework is representing Feynman diagrams as computational graphs, which structures the inherent mathematical complexity and facilitates the application of optimized algorithms developed for machine learning and high-performance computing. Consequently, automatic differentiation, native to these graph representations, delivers efficient, fully automated, high-order field-theoretic renormalization procedures. This graph-centric approach also enables sophisticated numerical integration; our neural-network-enhanced Monte Carlo method, accelerated via massively parallel GPU implementation, efficiently evaluates challenging high-dimensional diagrammatic integrals. Applying this framework to the uniform electron gas, we determine the quasiparticle effective mass to a precision significantly surpassing current state-of-the-art simulations. Our work demonstrates the transformative potential of integrating AI-driven computational advances with QFT, opening systematic pathways for solving complex quantum many-body problems across disciplines.

hep-th↗