arXiv subjects

Siyu Wang

Publications and source records attributed to Siyu Wang.

At least 37 records · Page 2Linked to original sources

arXiv2025

Effective Attention-Guided Multi-Scale Medical Network for Skin Lesion Segmentation

In the field of healthcare, precise skin lesion segmentation is crucial for the early detection and accurate diagnosis of skin diseases. Despite significant advances in deep learning for image processing, existing methods have yet to effectively address the challenges of irregular lesion shapes and low contrast. To address these issues, this paper proposes an innovative encoder-decoder network architecture based on multi-scale residual structures, capable of extracting rich feature information from different receptive fields to effectively identify lesion areas. By introducing a Multi-Resolution Multi-Channel Fusion (MRCF) module, our method captures cross-scale features, enhancing the clarity and accuracy of the extracted information. Furthermore, we propose a Cross-Mix Attention Module (CMAM), which redefines the attention scope and dynamically calculates weights across multiple contexts, thus improving the flexibility and depth of feature capture and enabling deeper exploration of subtle features. To overcome the information loss caused by skip connections in traditional U-Net, an External Attention Bridge (EAB) is introduced, facilitating the effective utilization of information in the decoder and compensating for the loss during upsampling. Extensive experimental evaluations on several skin lesion segmentation datasets demonstrate that the proposed model significantly outperforms existing transformer and convolutional neural network-based models, showcasing exceptional segmentation accuracy and robustness.

cs.CV↗

arXiv2025

STCTS: Generative Semantic Compression for Ultra-Low Bitrate Speech via Explicit Text-Prosody-Timbre Decomposition

Voice communication in bandwidth-constrained environments--maritime, satellite, and tactical networks--remains prohibitively expensive. Traditional codecs struggle below 1 kbps, while existing semantic approaches (STT-TTS) sacrifice prosody and speaker identity. We present STCTS, a generative semantic compression framework enabling natural voice communication at 80 bps. STCTS explicitly decomposes speech into linguistic content, prosodic expression, and speaker timbre, applying tailored compression: context-aware text encoding (70 bps), sparse prosody transmission via TTS interpolation (<14 bps at 0.1-1 Hz), and amortized speaker embedding. Evaluations on LibriSpeech demonstrate a 75x bitrate reduction versus Opus (6 kbps) and 12x versus EnCodec (1 kbps), while maintaining perceptual quality (NISQA MOS > 4.26), graceful degradation under packet loss and noise resilience. We also discover a bimodal quality distribution with prosody sampling rate: sparse and dense updates both achieve high quality, while mid-range rates degrade due to perceptual discontinuities--guiding optimal configuration design. Beyond efficiency, our modular architecture supports privacy-preserving encryption, human-interpretable transmission, and flexible deployment on edge devices, offering a robust solution for ultra-low bandwidth scenarios.

cs.SD↗

arXiv2025

Energy-Guided Diffusion Sampling for Long-Term User Behavior Prediction in Reinforcement Learning-based Recommendation

Reinforcement learning-based recommender systems (RL4RS) have gained attention for their ability to adapt to dynamic user preferences. However, these systems face challenges, particularly in offline settings, where data inefficiency and reliance on pre-collected trajectories limit their broader applicability. While offline reinforcement learning methods leverage extensive datasets to address these issues, they often struggle with noisy data and fail to capture long-term user preferences, resulting in suboptimal recommendation policies. To overcome these limitations, we propose Diffusion-enhanced Actor-Critic for Offline RL4RS (DAC4Rec), a novel framework that integrates diffusion processes with reinforcement learning to model complex user preferences more effectively. DAC4Rec leverages the denoising capabilities of diffusion models to enhance the robustness of offline RL algorithms and incorporates a Q-value-guided policy optimization strategy to better handle suboptimal trajectories. Additionally, we introduce an energy-based sampling strategy to reduce randomness during recommendation generation, ensuring more targeted and reliable outcomes. We validate the effectiveness of DAC4Rec through extensive experiments on six real-world offline datasets and in an online simulation environment, demonstrating its ability to optimize long-term user preferences. Furthermore, we show that the proposed diffusion policy can be seamlessly integrated into other commonly used RL algorithms in RL4RS, highlighting its versatility and wide applicability.

cs.IR↗

arXiv2025

Maximum In-Support Return Modeling for Dynamic Recommendation with Language Model Prior

Reinforcement Learning-based recommender systems (RLRS) offer an effective way to handle sequential recommendation tasks but often face difficulties in real-world settings, where user feedback data can be sub-optimal or sparse. In this paper, we introduce MDT4Rec, an offline RLRS framework that builds on the Decision Transformer (DT) to address two major challenges: learning from sub-optimal histories and representing complex user-item interactions. First, MDT4Rec shifts the trajectory stitching procedure from the training phase to action inference, allowing the system to shorten its historical context when necessary and thereby ignore negative or unsuccessful past experiences. Second, MDT4Rec initializes DT with a pre-trained large language model (LLM) for knowledge transfer, replaces linear embedding layers with Multi-Layer Perceptrons (MLPs) for more flexible representations, and employs Low-Rank Adaptation (LoRA) to efficiently fine-tune only a small subset of parameters. We evaluate MDT4Rec on five public datasets and in an online simulation environment, demonstrating that it outperforms existing methods.

cs.IR↗

arXiv2025

On the statistical physics and thermodynamics of polymer networks: a Eulerian theory for entropic elasticity

This study presents a Eulerian theory to elucidate the molecular kinematics in polymer networks and their connection to continuum deformation, grounded in fundamental statistical physics and thermodynamics. Three key innovations are incorporated: 1. The network behavior is described through a global thermodynamic equilibrium condition that maximizes the number of accessible microstates for all segments, instead of directly dealing with the well-established single-chain models commonly adopted in traditional approaches. A variational problem is then posed in the Eulerian framework to identify this equilibrium state under geometric fluctuation constraints. Its solution recaptures the classical single-chain model and reveals the dependence of chain kinematics upon continuum deformation. 2. The chain stretch and orientation probability are found to be explicitly specified through the Eulerian logarithmic strain and spatial direction. The resulting hyperelastic model, with only two physical parameters, outperforms the extant models with same number of parameters. It further provides a physical justification for prior models exhibiting superior predictive capabilities: the model becomes equivalent to the Biot-chain model at moderate deformations, while converging to the classical Hencky strain energy in the small strain limit. 3.A novel biaxial instability emerges as a phase transition in chain orientation. At sufficiently large deformation, chains increasingly align with the primary stretched direction, depleting their density in other directions. Consequently, the stresses in non-primary stretched directions would decrease as the loss in chain density outweighs the gain in chain force. For equal biaxial tension, instability is therefore triggered because perfect equality of the two principal stretches without any perturbation is practically unachievable.

cond-mat.soft↗

arXiv2025

A statistical theory for polymer elasticity: from molecular kinematics to continuum behavior

Predicting the macroscopic mechanical behavior of polymeric materials from the micro-structural features has remained a challenge for decades. Existing theoretical models often fail to accurately capture the experimental data, due to non-physical assumptions that link the molecule kinematics with the macroscopic deformation. In this work, we construct a novel Hamiltonian for chain segments enabling a unified statistical description of both individual macromolecular chains and continuum polymer networks. The chain kinematics, including the stretch and orientation properties, are retrieved by the thermodynamic observables without phenomenological assumptions. The theory shows that the chain stretch is specified by a simple relation via its current spatial direction and the continuum Eulerian logarithmic strain, while the probability of a chain in this spatial direction is governed by the new Hamiltonian of a single segment. The model shows a significantly improved prediction on the hyperelastic response of elastomers, relying on minimal, physically-grounded parameters.

cond-mat.soft↗

arXiv2025

Diffusion Policies for Risk-Averse Behavior Modeling in Offline Reinforcement Learning

Offline reinforcement learning (RL) presents distinct challenges as it relies solely on observational data. A central concern in this context is ensuring the safety of the learned policy by quantifying uncertainties associated with various actions and environmental stochasticity. Traditional approaches primarily emphasize mitigating epistemic uncertainty by learning risk-averse policies, often overlooking environmental stochasticity. In this study, we propose an uncertainty-aware distributional offline RL method to simultaneously address both epistemic uncertainty and environmental stochasticity. We propose a model-free offline RL algorithm capable of learning risk-averse policies and characterizing the entire distribution of discounted cumulative rewards, as opposed to merely maximizing the expected value of accumulated discounted returns. Our method is rigorously evaluated through comprehensive experiments in both risk-sensitive and risk-neutral benchmarks, demonstrating its superior performance.

cs.LG↗

arXiv2025

Lightweight Fingernail Haptic Device: Unobstructed Fingerpad Force and Vibration Feedback for Enhanced Virtual Dexterous Manipulation

This study presents a lightweight, wearable fingertip haptic device that provides physics-based haptic feedback for dexterous manipulation in virtual environments without hindering real-world interactions. The device, designed with thin strings and actuators attached to the fingernails, ensures minimal weight (1.55 g per finger) and preserves finger flexibility. Integrating the software with a physics engine renders multiple types of haptic feedback (grip force, collision, and sliding vibration feedback). We evaluated the device's performance in pressure perception, slip feedback, typical dexterous manipulation tasks, and daily operations, and we gathered user experience through subjective assessments. Our results show that participants could perceive and respond to pressure and vibration feedback. Through dexterous manipulation experiments, we further demonstrated that these minimal haptic cues significantly improved virtual task efficiency, showcasing how lightweight haptic feedback can enhance manipulation performance without complex mechanisms. The device's ability to preserve tactile sensations and minimize hindrance to real-world operations is a key advantage over glove-type haptic devices. This research offers a potential solution for designing haptic interfaces that balance lightweight construction, haptic feedback for dexterous manipulation, and daily wearability.

cs.HC↗

arXiv2025

CAD-GPT: Synthesising CAD Construction Sequence with Spatial Reasoning-Enhanced Multimodal LLMs

Computer-aided design (CAD) significantly enhances the efficiency, accuracy, and innovation of design processes by enabling precise 2D and 3D modeling, extensive analysis, and optimization. Existing methods for creating CAD models rely on latent vectors or point clouds, which are difficult to obtain, and storage costs are substantial. Recent advances in Multimodal Large Language Models (MLLMs) have inspired researchers to use natural language instructions and images for CAD model construction. However, these models still struggle with inferring accurate 3D spatial location and orientation, leading to inaccuracies in determining the spatial 3D starting points and extrusion directions for constructing geometries. This work introduces CAD-GPT, a CAD synthesis method with spatial reasoning-enhanced MLLM that takes either a single image or a textual description as input. To achieve precise spatial inference, our approach introduces a 3D Modeling Spatial Mechanism. This method maps 3D spatial positions and 3D sketch plane rotation angles into a 1D linguistic feature space using a specialized spatial unfolding mechanism, while discretizing 2D sketch coordinates into an appropriate planar space to enable precise determination of spatial starting position, sketch orientation, and 2D sketch coordinate translations. Extensive experiments demonstrate that CAD-GPT consistently outperforms existing state-of-the-art methods in CAD model synthesis, both quantitatively and qualitatively.

cs.CV↗

arXiv2025

Debiased distributed PCA under high dimensional spiked model

We study distributed principal component analysis (PCA) in high-dimensional settings under the spiked model. In such regimes, sample eigenvectors can deviate significantly from population ones, introducing a persistent bias. Existing distributed PCA methods are sensitive to this bias, particularly when the number of machines is small. Their consistency typically relies on the number of machines tending to infinity. We propose a debiased distributed PCA algorithm that corrects the local bias before aggregation and incorporates a sparsity-detection step to adaptively handle sparse and non-sparse eigenvectors. Theoretically, we establish the consistency of our estimator under much weaker conditions compared to existing literature. In particular, our approach does not require symmetric innovations and only assumes a finite sixth moment. Furthermore, our method generally achieves smaller estimation error, especially when the number of machines is small. Empirically, extensive simulations and real data experiments demonstrate that our method consistently outperforms existing distributed PCA approaches. The advantage is especially prominent when the leading eigenvectors are sparse or the number of machines is limited. Our method and theoretical analysis are also applicable to the sample correlation matrix.

stat.ME↗

arXiv2025

HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformer

Recent advancements in image generative foundation models have prioritized quality improvements but often at the cost of increased computational complexity and inference latency. To address this critical trade-off, we introduce HiDream-I1, a new open-source image generative foundation model with 17B parameters that achieves state-of-the-art image generation quality within seconds. HiDream-I1 is constructed with a new sparse Diffusion Transformer (DiT) structure. Specifically, it starts with a dual-stream decoupled design of sparse DiT with dynamic Mixture-of-Experts (MoE) architecture, in which two separate encoders are first involved to independently process image and text tokens. Then, a single-stream sparse DiT structure with dynamic MoE architecture is adopted to trigger multi-model interaction for image generation in a cost-efficient manner. To support flexiable accessibility with varied model capabilities, we provide HiDream-I1 in three variants: HiDream-I1-Full, HiDream-I1-Dev, and HiDream-I1-Fast. Furthermore, we go beyond the typical text-to-image generation and remould HiDream-I1 with additional image conditions to perform precise, instruction-based editing on given images, yielding a new instruction-based image editing model namely HiDream-E1. Ultimately, by integrating text-to-image generation and instruction-based image editing, HiDream-I1 evolves to form a comprehensive image agent (HiDream-A1) capable of fully interactive image creation and refinement. To accelerate multi-modal AIGC research, we have open-sourced all the codes and model weights of HiDream-I1-Full, HiDream-I1-Dev, HiDream-I1-Fast, HiDream-E1 through our project websites: https://github.com/HiDream-ai/HiDream-I1 and https://github.com/HiDream-ai/HiDream-E1. All features can be directly experienced via https://vivago.ai/studio.

cs.CV↗

arXiv2025

A New Forward Discriminant Analysis Framework Based On Pillai's Trace and ULDA

Linear discriminant analysis (LDA), a traditional classification tool, suffers from limitations such as sensitivity to noise and computational challenges when dealing with non-invertible within-class scatter matrices. Traditional stepwise LDA frameworks, which iteratively select the most informative features, often exacerbate these issues by relying heavily on Wilks' $Λ$, potentially causing premature stopping of the selection process. This paper introduces a novel forward discriminant analysis framework that integrates Pillai's trace with Uncorrelated Linear Discriminant Analysis (ULDA) to address these challenges, and offers a unified and stand-alone classifier. Through simulations and real-world datasets, the new framework demonstrates effective control of Type I error rates and improved classification accuracy, particularly in cases involving perfect group separations. The results highlight the potential of this approach as a robust alternative to the traditional stepwise LDA framework.

stat.ME↗

arXiv2025

FoLDTree: A ULDA-Based Decision Tree Framework for Efficient Oblique Splits and Feature Selection

Traditional decision trees are limited by axis-orthogonal splits, which can perform poorly when true decision boundaries are oblique. While oblique decision tree methods address this limitation, they often face high computational costs, difficulties with multi-class classification, and a lack of effective feature selection. In this paper, we introduce LDATree and FoLDTree, two novel frameworks that integrate Uncorrelated Linear Discriminant Analysis (ULDA) and Forward ULDA into a decision tree structure. These methods enable efficient oblique splits, handle missing values, support feature selection, and provide both class labels and probabilities as model outputs. Through evaluations on simulated and real-world datasets, LDATree and FoLDTree consistently outperform axis-orthogonal and other oblique decision tree methods, achieving accuracy levels comparable to the random forest. The results highlight the potential of these frameworks as robust alternatives to traditional single-tree methods.

cs.LG↗

arXiv2025

Fast microwave-driven two-qubit gates between fluxonium qubits with a transmon coupler

Two qubit gates constitute fundamental building blocks in the realization of large-scale quantum devices. Using superconducting circuits, two-qubit gates have previously been implemented in different ways with each method aiming to maximize gate fidelity. Another important goal of a new gate scheme is to minimize the complexity of gate calibration. In this work, we demonstrate a high-fidelity two-qubit gate between two fluxonium qubits enabled by an intermediate capacitively coupled transmon. The coupling strengths between the qubits and the coupler are designed to minimize residual crosstalk while still allowing for fast gate operations. The gate is based on frequency selectively exciting the coupler using a microwave drive to complete a 2$π$ rotation, conditional on the state of the fluxonium qubits. When successful, this drive scheme implements a conditional phase gate. Using analytically derived pulse shapes, we minimize unwanted excitations of the coupler and obtain gate errors of $10^{-2}$ for gate times below 60~ns. At longer durations, our gate is limited by relaxation of the coupler. Our results show how carefully designed control pulses can speed up frequency selective entangling gates.

quant-ph↗

arXiv2025

Enhancing Embedding Representation Stability in Recommendation Systems with Semantic ID

The exponential growth of online content has posed significant challenges to ID-based models in industrial recommendation systems, ranging from extremely high cardinality and dynamically growing ID space, to highly skewed engagement distributions, to prediction instability as a result of natural id life cycles (e.g, the birth of new IDs and retirement of old IDs). To address these issues, many systems rely on random hashing to handle the id space and control the corresponding model parameters (i.e embedding table). However, this approach introduces data pollution from multiple ids sharing the same embedding, leading to degraded model performance and embedding representation instability. This paper examines these challenges and introduces Semantic ID prefix ngram, a novel token parameterization technique that significantly improves the performance of the original Semantic ID. Semantic ID prefix ngram creates semantically meaningful collisions by hierarchically clustering items based on their content embeddings, as opposed to random assignments. Through extensive experimentation, we demonstrate that Semantic ID prefix ngram not only addresses embedding instability but also significantly improves tail id modeling, reduces overfitting, and mitigates representation shifts. We further highlight the advantages of Semantic ID prefix ngram in attention-based models that contextualize user histories, showing substantial performance improvements. We also report our experience of integrating Semantic ID into Meta production Ads Ranking system, leading to notable performance gains and enhanced prediction stability in live deployments.

cs.IR↗

arXiv2025

Spatial Analysis of Neuromuscular Junctions Activation in Three-Dimensional Histology-based Muscle Reconstructions

Histology has long been a foundational technique for studying anatomical structures through tissue slicing. Advances in computational methods now enable three dimensional (3D) reconstruction of organs from histology images, enhancing the analysis of structural and functional features. Here, we present a novel multimodal computational method to reconstruct rodent muscles in 3D using classical image processing and data analysis techniques, analyze their structural features and correlate them to previously recorded electrophysiological data. The algorithm analyzes spatial distribution patterns of features identified through histological staining, normalizing them across multiple samples. Further, the algorithm successfully correlates spatial patterns with high density epimysial ElectroMyoGraphy (hdEMG) recordings, providing a multimodal perspective on neuromuscular dynamics, linking spatial and electrophysiological information. The code was validated by looking at the distribution of NeuroMuscular Junctions (NMJs) in naive soleus muscles and compared the distributions and patterns observed with ones observed in previous literature. Our results showed consistency with the expected results, validating our method for features and pattern recognition. The multimodal aspect was shown in a naive soleus muscle, where a strong correlation was found between motor unit locations derived via hdEMG, and NMJ locations obtained from histology, highlighting their spatial relationship. This multimodal analysis tool integrates 3D structural data with electrophysiological activity, opening new avenues in muscle diagnostics, regenerative medicine, and personalized therapies where spatial insights could one day predict electrophysiological behavior or vice versa.

q-bio.QM↗

arXiv2025

Policy-Guided Causal State Representation for Offline Reinforcement Learning Recommendation

In offline reinforcement learning-based recommender systems (RLRS), learning effective state representations is crucial for capturing user preferences that directly impact long-term rewards. However, raw state representations often contain high-dimensional, noisy information and components that are not causally relevant to the reward. Additionally, missing transitions in offline data make it challenging to accurately identify features that are most relevant to user satisfaction. To address these challenges, we propose Policy-Guided Causal Representation (PGCR), a novel two-stage framework for causal feature selection and state representation learning in offline RLRS. In the first stage, we learn a causal feature selection policy that generates modified states by isolating and retaining only the causally relevant components (CRCs) while altering irrelevant components. This policy is guided by a reward function based on the Wasserstein distance, which measures the causal effect of state components on the reward and encourages the preservation of CRCs that directly influence user interests. In the second stage, we train an encoder to learn compact state representations by minimizing the mean squared error (MSE) loss between the latent representations of the original and modified states, ensuring that the representations focus on CRCs. We provide a theoretical analysis proving the identifiability of causal effects from interventions, validating the ability of PGCR to isolate critical state components for decision-making. Extensive experiments demonstrate that PGCR significantly improves recommendation performance, confirming its effectiveness for offline RL-based recommender systems.

cs.IR↗

arXiv2025

Agent Hospital: A Simulacrum of Hospital with Evolvable Medical Agents

The recent rapid development of large language models (LLMs) has sparked a new wave of technological revolution in medical artificial intelligence (AI). While LLMs are designed to understand and generate text like a human, autonomous agents that utilize LLMs as their "brain" have exhibited capabilities beyond text processing such as planning, reflection, and using tools by enabling their "bodies" to interact with the environment. We introduce a simulacrum of hospital called Agent Hospital that simulates the entire process of treating illness, in which all patients, nurses, and doctors are LLM-powered autonomous agents. Within the simulacrum, doctor agents are able to evolve by treating a large number of patient agents without the need to label training data manually. After treating tens of thousands of patient agents in the simulacrum (human doctors may take several years in the real world), the evolved doctor agents outperform state-of-the-art medical agent methods on the MedQA benchmark comprising US Medical Licensing Examination (USMLE) test questions. Our methods of simulacrum construction and agent evolution have the potential in benefiting a broad range of applications beyond medical AI.

cs.AI↗