arXiv subjects

Sheng Liu

Publications and source records attributed to Sheng Liu.

At least 73 records · Page 4Linked to original sources

arXiv2025

MedTrinity-25M: A Large-scale Multimodal Dataset with Multigranular Annotations for Medicine

This paper introduces MedTrinity-25M, a comprehensive, large-scale multimodal dataset for medicine, covering over 25 million images across 10 modalities with multigranular annotations for more than 65 diseases. These multigranular annotations encompass both global information, such as modality and organ detection, and local information like ROI analysis, lesion texture, and region-wise correlations. Unlike the existing multimodal datasets, which are limited by the availability of image-text pairs, we have developed the first automated pipeline that scales up multimodal data by generating multigranular visual and textual annotations in the form of image-ROI-description triplets without the need for any paired text descriptions. Specifically, data from over 30 different sources have been collected, preprocessed, and grounded using domain-specific expert models to identify ROIs related to abnormal regions. We then build a comprehensive knowledge base and prompt multimodal large language models to perform retrieval-augmented generation with the identified ROIs as guidance, resulting in multigranular textual descriptions. Compared to existing datasets, MedTrinity-25M provides the most enriched annotations, supporting a comprehensive range of multimodal tasks such as captioning and report generation, as well as vision-centric tasks like classification and segmentation. We propose LLaVA-Tri by pretraining LLaVA on MedTrinity-25M, achieving state-of-the-art performance on VQA-RAD, SLAKE, and PathVQA, surpassing representative SOTA multimodal large language models. Furthermore, MedTrinity-25M can also be utilized to support large-scale pre-training of multimodal medical AI models, contributing to the development of future foundation models in the medical domain. We will make our dataset available.

cs.CV↗

arXiv2025

DESign: Dynamic Context-Aware Convolution and Efficient Subnet Regularization for Continuous Sign Language Recognition

Current continuous sign language recognition (CSLR) methods struggle with handling diverse samples. Although dynamic convolutions are ideal for this task, they mainly focus on spatial modeling and fail to capture the temporal dynamics and contextual dependencies. To address this, we propose DESign, a novel framework that incorporates Dynamic Context-Aware Convolution (DCAC) and Subnet Regularization Connectionist Temporal Classification (SR-CTC). DCAC dynamically captures the inter-frame motion cues that constitute signs and uniquely adapts convolutional weights in a fine-grained manner based on contextual information, enabling the model to better generalize across diverse signing behaviors and boost recognition accuracy. Furthermore, we observe that existing methods still rely on only a limited number of frames for parameter updates during training, indicating that CTC learning overfits to a dominant path. To address this, SR-CTC regularizes training by applying supervision to subnetworks, encouraging the model to explore diverse CTC alignment paths and effectively preventing overfitting. A classifier-sharing strategy in SR-CTC further strengthens multi-scale consistency. Notably, SR-CTC introduces no inference overhead and can be seamlessly integrated into existing CSLR models to boost performance. Extensive ablations and visualizations further validate the effectiveness of the proposed methods. Results on mainstream CSLR datasets (i.e., PHOENIX14, PHOENIX14-T, CSL-Daily) demonstrate that DESign achieves state-of-the-art performance.

cs.CV↗

arXiv2025

More Thinking, Less Seeing? Assessing Amplified Hallucination in Multimodal Reasoning Models

Test-time compute has empowered multimodal large language models to generate extended reasoning chains, yielding strong performance on tasks such as multimodal math reasoning. However, this improved reasoning ability often comes with increased hallucination: as generations become longer, models tend to drift away from image-grounded content and rely more heavily on language priors. Attention analysis shows that longer reasoning chains lead to reduced focus on visual inputs, which contributes to hallucination. To systematically study this phenomenon, we introduce RH-AUC, a metric that quantifies how a model's perception accuracy changes with reasoning length, allowing us to evaluate whether the model preserves visual grounding during reasoning. We also release RH-Bench, a diagnostic benchmark that spans a variety of multimodal tasks, designed to assess the trade-off between reasoning ability and hallucination. Our analysis reveals that (i) larger models typically achieve a better balance between reasoning and perception, and (ii) this balance is influenced more by the types and domains of training data than by its overall volume. These findings underscore the importance of evaluation frameworks that jointly consider both reasoning quality and perceptual fidelity.

cs.CL↗

arXiv2025

Knowledge or Reasoning? A Close Look at How LLMs Think Across Domains

Recent advances in reasoning-enhanced Large Language Models such as OpenAI-o1/3 and DeepSeek-R1 have significantly improved performance on complex tasks. However, the quality and transparency of their internal reasoning processes remain underexplored. This work moves beyond the final-answer accuracy and investigates step-by-step reasoning in the medical and mathematical domains by explicitly decomposing the thinking trajectories into two parts: knowledge and reasoning. Specifically, we introduce a fine-grained evaluation framework that judges: (1) the correctness of knowledge used (measured by Knowledge Index (KI)) and (2) the quality of reasoning (measured by Information Gain (InfoGain)). Using this framework, we study R1-distilled and base Qwen models trained with supervised fine-tuning (SFT) and/or reinforcement learning (RL) in the medical and math domains. Three intriguing findings emerge: (1) The general reasoning abilities in R1-distilled models do not transfer effectively to the medical domain through either SFT or RL. (2) SFT raises final-answer accuracy in both domains, but often at the cost of reasoning quality: InfoGain drops by 38.9% on average compared with untrained models; In the medical domain, however, SFT remains crucial because domain knowledge is indispensable. (3) RL enhances medical reasoning by pruning inaccurate or irrelevant knowledge from reasoning paths, thereby improving both reasoning accuracy and knowledge correctness.

cs.CL↗

arXiv2025

Spontaneous Enhancement of Dzyaloshinskii-Moriya Interaction via Field-Cooling-Induced Interface Engineering in 2D van der Waals Ferromagnetic ternary Tellurides

The emergence of two-dimensional (2D) van der Waals (vdW) ferromagnets has opened new avenues for exploring topological spin textures and their applications in next-generation spintronics. Among these materials, Fe3GaTe2 (FGaT) emerges as a model system due to its room-temperature skyrmion phases, which are stabilized by strong Dzyaloshinskii-Moriya interaction (DMI). However, the atomistic origins of DMI in centrosymmetric vdW lattices remain elusive. Here, we report a spontaneous DMI enhancement mechanism driven by FC in FGaT and its analog Fe3GeTe2 (FGeT). Combining Raman spectroscopy and scanning transmission electron microscopy (STEM), we have observed the irreversible precipitation of FeTe2 in annealed FGaT. The resulting FeTe2/FGaT heterostructure is considered to break the symmetry and significantly enhance the DMI. Furthermore, similar phenomenon has been observed in the family ferromagnetic material FGeT as well. Additionally, the precipitation of FeTe2 varies significantly with different thicknesses of FGaT, aligning closely with the reported behavior of skyrmions. This discovery provides new insights into the mechanisms behind the origin of the DMI in ternary tellurides, paving the way for advanced spintronic applications.

cond-mat.mtrl-sci↗

arXiv2025

Strain Induced Robust Skyrmion lattice at Room Temperature in van der Waals Ferromagnet

Manipulating topological magnetic orders of two-dimensional (2D) magnets by strain, once achieved, offers enormous potential for future low-power flexible spintronic applications. In this work, by placing Fe3GaTe2 (FGaT), a room-temperature 2D ferromagnet, on flexible substrate, we demonstrate a field-free and robust formation of skyrmion lattice induced by strain. By applying a minimal strain of ~0.80% to pre-annealed FGaT flakes, the Magnetic Force Microscopy (MFM) tip directly triggers the transition from maze-like domains to an ordered skyrmion lattice while scanning the sample surface. The skyrmion lattice is rather stable against extensive cyclic mechanical testing (stretching, bending, and twisting over 2000 cycles each). It also exhibited stability across a wide range of magnetic fields (~2.9 kOe) and temperatures (~ 323 K), as well as long-term retention stability, highlighting its robustness and field free stabilization. The strain effect reduces the lattice symmetry and enhances the Dzyaloshinskii-Moriya interaction (DMI) of FGaT, thus stabilizing the skyrmion lattice. Our findings highlight the potential of FGaT for integrating magnetic skyrmions into future low-power-consumption flexible spintronics devices.

cond-mat.mtrl-sci↗

arXiv2025

On-demand longitudinal transformations of light beams via an amplitude-phase-decoupled Pancharatnam-Berry phase element

Overcoming the inherent phase-only modulation limitation of conventional Pancharatnam-Berry (PB) elements, we propose a checkerboard-encoded PB element that enables simultaneous amplitude and phase modulation through PB phase engineering. This single-element platform directly generates polarization-rotating beams with continuously varying polarization angles along the propagation axis, eliminating traditional dynamic phase requirements. By strategically combining the proposed element with a liquid crystal q-plate and waveplates, we demonstrate on-demand longitudinal transformations of polarization states and high-order Poincaré sphere beams. Significantly, we achieve, for the first time to our knowledge, the propagation-transformed optical skyrmions, whose topological textures evolve with propagation distance. This amplitude-phase decoupling mechanism and the integration solution for various longitudinal transformations open new avenues in structured light manipulation, particularly benefiting quantum optics and topological photonics applications requiring longitudinal mode control.

physics.optics↗

arXiv2025

Prototype-Based Information Compensation Network for Multi-Source Remote Sensing Data Classification

Multi-source remote sensing data joint classification aims to provide accuracy and reliability of land cover classification by leveraging the complementary information from multiple data sources. Existing methods confront two challenges: inter-frequency multi-source feature coupling and inconsistency of complementary information exploration. To solve these issues, we present a Prototype-based Information Compensation Network (PICNet) for land cover classification based on HSI and SAR/LiDAR data. Specifically, we first design a frequency interaction module to enhance the inter-frequency coupling in multi-source feature extraction. The multi-source features are first decoupled into high- and low-frequency components. Then, these features are recoupled to achieve efficient inter-frequency communication. Afterward, we design a prototype-based information compensation module to model the global multi-source complementary information. Two sets of learnable modality prototypes are introduced to represent the global modality information of multi-source data. Subsequently, cross-modal feature integration and alignment are achieved through cross-attention computation between the modality-specific prototype vectors and the raw feature representations. Extensive experiments on three public datasets demonstrate the significant superiority of our PICNet over state-of-the-art methods. The codes are available at https://github.com/oucailab/PICNet.

eess.IV↗

arXiv2025

RoboBERT: An End-to-end Multimodal Robotic Manipulation Model

Embodied intelligence seamlessly integrates vision, language, and action.~However, most multimodal robotic models rely on massive fine-tuning, incurring high time and hardware costs.~To address this, we introduce RoboBERT, an end-to-end multimodal manipulation model built around a novel two-stage training paradigm.~In the first stage, we freeze most of the vision encoder and train with a single "standard" instruction phrasing, allowing the model to focus on stable policy learning via a CNN-based diffusion policy.~In the second stage, we unfreeze all modules and inject diverse natural language variants, rapidly aligning varied instructions to the already-learned policy without destabilizing performance.~We further employ systematic data augmentations to enhance robustness against visual perturbations.~Without relying on auxiliary datasets, RoboBERT achieves new state-of-the-art (SOTA) mean episode lengths of 4.52 on the CALVIN ABCD-D benchmark and 3.79 on the ABC-D benchmark using only language-labeled expert demonstrations and a comparatively lightweight architecture.Real-robot trials on a 6-DOF manipulator confirm higher success rates than comparable methods trained on identical data.These results demonstrate that our data-augmentation-enhanced two-stage training paradigm delivers efficient, scalable, and broadly applicable performance for multimodal robotic systems.

cs.RO↗

arXiv2025

Scalable multilayer diffractive neural network with all-optical nonlinear activation

All-optical diffractive neural networks (DNNs) offer a promising alternative to electronics-based neural network processing due to their low latency, high throughput, and inherent spatial parallelism. However, the lack of reconfigurability and nonlinearity limits existing all-optical DNNs to handling only simple tasks. In this study, we present a folded optical system that enables a multilayer reconfigurable DNN using a single spatial light modulator. This platform not only enables dynamic weight reconfiguration for diverse classification challenges but crucially integrates a mirror-coated silicon substrate exhibiting instantaneous \c{hi}(3) nonlinearity. The incorporation of all-optical nonlinear activation yields substantial accuracy improvements across benchmark tasks, with performance gains becoming increasingly significant as both network depth and task complexity escalate. Our system represents a critical advancement toward realizing scalable all-optical neural networks with complex architectures, potentially achieving computational capabilities that rival their electronic counterparts while maintaining photonic advantages.

physics.optics↗

arXiv2025

Automated radiotherapy treatment planning guided by GPT-4Vision

Objective: Radiotherapy treatment planning is a time-consuming and potentially subjective process that requires the iterative adjustment of model parameters to balance multiple conflicting objectives. Recent advancements in frontier Artificial Intelligence (AI) models offer promising avenues for addressing the challenges in planning and clinical decision-making. This study introduces GPT-RadPlan, an automated treatment planning framework that integrates radiation oncology knowledge with the reasoning capabilities of large multi-modal models, such as GPT-4Vision (GPT-4V) from OpenAI. Approach: Via in-context learning, we incorporate clinical requirements and a few (3 in our experiments) approved clinical plans with their optimization settings, enabling GPT-4V to acquire treatment planning domain knowledge. The resulting GPT-RadPlan system is integrated into our in-house inverse treatment planning system through an application programming interface (API). For a given patient, GPT-RadPlan acts as both plan evaluator and planner, first assessing dose distributions and dose-volume histograms (DVHs), and then providing textual feedback on how to improve the plan to match the physician's requirements. In this manner, GPT-RadPlan iteratively refines the plan by adjusting planning parameters, such as weights and dose objectives, based on its suggestions. Main results: The efficacy of the automated planning system is showcased across 17 prostate cancer and 13 head and neck cancer VMAT plans with prescribed doses of 70.2 Gy and 72 Gy, respectively, where we compared GPT-RadPlan results to clinical plans produced by human experts. In all cases, GPT-RadPlan either outperformed or matched the clinical plans, demonstrating superior target coverage and reducing organ-at-risk doses by 5 Gy on average (15 percent for prostate and 10-15 percent for head and neck).

physics.med-ph↗

arXiv2025

MedReason: Eliciting Factual Medical Reasoning Steps in LLMs via Knowledge Graphs

Medical tasks such as diagnosis and treatment planning require precise and complex reasoning, particularly in life-critical domains. Unlike mathematical reasoning, medical reasoning demands meticulous, verifiable thought processes to ensure reliability and accuracy. However, there is a notable lack of datasets that provide transparent, step-by-step reasoning to validate and enhance the medical reasoning ability of AI models. To bridge this gap, we introduce MedReason, a large-scale high-quality medical reasoning dataset designed to enable faithful and explainable medical problem-solving in large language models (LLMs). We utilize a structured medical knowledge graph (KG) to convert clinical QA pairs into logical chains of reasoning, or ``thinking paths'', which trace connections from question elements to answers via relevant KG entities. Each path is validated for consistency with clinical logic and evidence-based medicine. Our pipeline generates detailed reasoning for various medical questions from 7 medical datasets, resulting in a dataset of 32,682 question-answer pairs, each with detailed, step-by-step explanations. Experiments demonstrate that fine-tuning with our dataset consistently boosts medical problem-solving capabilities, achieving significant gains of up to 7.7% for DeepSeek-Ditill-8B. Our top-performing model, MedReason-8B, outperforms the Huatuo-o1-8B, a state-of-the-art medical reasoning model, by up to 4.2% on the clinical benchmark MedBullets. We also engage medical professionals from diverse specialties to assess our dataset's quality, ensuring MedReason offers accurate and coherent medical reasoning. Our data, models, and code is available at https://github.com/UCSC-VLAA/MedReason.

cs.CL↗

arXiv2025

Data Reconstruction Attacks and Defenses: A Systematic Evaluation

Reconstruction attacks and defenses are essential in understanding the data leakage problem in machine learning. However, prior work has centered around empirical observations of gradient inversion attacks, lacks theoretical grounding, and cannot disentangle the usefulness of defending methods from the computational limitation of attacking methods. In this work, we propose to view the problem as an inverse problem, enabling us to theoretically and systematically evaluate the data reconstruction attack. On various defense methods, we derived the algorithmic upper bound and the matching (in feature dimension and architecture dimension) information-theoretical lower bound on the reconstruction error for two-layer neural networks. To complement the theoretical results and investigate the utility-privacy trade-off, we defined a natural evaluation metric of the defense methods with similar utility loss among the strongest attacks. We further propose a strong reconstruction attack that helps update some previous understanding of the strength of defense methods under our proposed evaluation metric.

cs.CR↗

arXiv2025

OLMD: Orientation-aware Long-term Motion Decoupling for Continuous Sign Language Recognition

The primary challenge in continuous sign language recognition (CSLR) mainly stems from the presence of multi-orientational and long-term motions. However, current research overlooks these crucial aspects, significantly impacting accuracy. To tackle these issues, we propose a novel CSLR framework: Orientation-aware Long-term Motion Decoupling (OLMD), which efficiently aggregates long-term motions and decouples multi-orientational signals into easily interpretable components. Specifically, our innovative Long-term Motion Aggregation (LMA) module filters out static redundancy while adaptively capturing abundant features of long-term motions. We further enhance orientation awareness by decoupling complex movements into horizontal and vertical components, allowing for motion purification in both orientations. Additionally, two coupling mechanisms are proposed: stage and cross-stage coupling, which together enrich multi-scale features and improve the generalization capabilities of the model. Experimentally, OLMD shows SOTA performance on three large-scale datasets: PHOENIX14, PHOENIX14-T, and CSL-Daily. Notably, we improved the word error rate (WER) on PHOENIX14 by an absolute 1.6% compared to the previous SOTA

cs.CV↗

arXiv2024

Enhancement of piezoelectric response in V doped LiNbO3 films deposited by RF magnetron sputtering

LiNbO3 films doped with vanadium (V) were deposited using RF magnetron sputtering technique. To realize doping with a wider range of V concentration, a 30 mm V metal inlaid target asymmetrically embedded in the 150 mm lithium niobate target was used. The V concentration in the deposited films was a decreasing function of the distance from the V target. The V/Nb ratio in the film decreased from 0.155 to 0.024. Surface and inner morphology and structure, phase and element composition, microstructure, and ferroelectric properties of the undoped and V doped LiNbO3 films were studied. The measured maximal d33 constant of the LiNbVO film with V/Nb ratio of 0.07 was about three times higher than that of the undoped LiNbO3 film, 13.5 pC/N and 4.76 pC/N, respectively. The optimal composition in the deposition geometry used was within the V/Nb ratio range of 0.05 to 0.13. Undoped and V doped LiNbO3 thin films were used as bulk acoustic wave ultrasonic transducers deposited on stainless steel plates to generate longitudinal waves and compare their ultrasonic performance.

cond-mat.mtrl-sci↗

arXiv2024

Dynamics of Topological Defects in a Rashba Spin-Orbit Coupled Bose-Einstein Condensate

We investigate the quench dynamics of a two-dimensional Rashba spin-orbit coupled Bose-Einstein condensate. Our study focuses on quenching the system from a zero-momentum phase to a plane-wave phase. During this quench, topological defects emerge in the form of vortices. These vortices and anti-vortices exhibit a random spatial distribution with equal numbers, mirroring the core principles of Kosterlitz-Thouless physics. In a uniform system, we observe an exponential scaling of both the vortex production time and the vortex number with the quench rate, consistent with the conventional Kibble-Zurek mechanism. The decay of which adheres to a logarithmic law, aligning with experimental observations.

cond-mat.quant-gas↗

arXiv2024

OTLRM: Orthogonal Learning-based Low-Rank Metric for Multi-Dimensional Inverse Problems

In real-world scenarios, complex data such as multispectral images and multi-frame videos inherently exhibit robust low-rank property. This property is vital for multi-dimensional inverse problems, such as tensor completion, spectral imaging reconstruction, and multispectral image denoising. Existing tensor singular value decomposition (t-SVD) definitions rely on hand-designed or pre-given transforms, which lack flexibility for defining tensor nuclear norm (TNN). The TNN-regularized optimization problem is solved by the singular value thresholding (SVT) operator, which leverages the t-SVD framework to obtain the low-rank tensor. However, it is quite complicated to introduce SVT into deep neural networks due to the numerical instability problem in solving the derivatives of the eigenvectors. In this paper, we introduce a novel data-driven generative low-rank t-SVD model based on the learnable orthogonal transform, which can be naturally solved under its representation. Prompted by the linear algebra theorem of the Householder transformation, our learnable orthogonal transform is achieved by constructing an endogenously orthogonal matrix adaptable to neural networks, optimizing it as arbitrary orthogonal matrices. Additionally, we propose a low-rank solver as a generalization of SVT, which utilizes an efficient representation of generative networks to obtain low-rank structures. Extensive experiments highlight its significant restoration enhancements.

cs.CV↗

arXiv2024

Learning Robust Grasping Strategy Through Tactile Sensing and Adaption Skill

Robust grasping represents an essential task in robotics, necessitating tactile feedback and reactive grasping adjustments for robust grasping of objects. Previous research has extensively combined tactile sensing with grasping, primarily relying on rule-based approaches, frequently neglecting post-grasping difficulties such as external disruptions or inherent uncertainties of the object's physics and geometry. To address these limitations, this paper introduces an human-demonstration-based adaptive grasping policy base on tactile, which aims to achieve robust gripping while resisting disturbances to maintain grasp stability. Our trained model generalizes to daily objects with seven different sizes, shapes, and textures. Experimental results demonstrate that our method performs well in dynamic and force interaction tasks and exhibits excellent generalization ability.

cs.RO↗