arXiv subjects

Ke Chen

Publications and source records attributed to Ke Chen.

At least 127 records · Page 7Linked to original sources

arXiv2025

Three-Dimensional Continuous Multi-Walled Carbon Nanotubes Network-Toughened Diamond Composite

Enhancing the fracture toughness of diamond while preserving its hardness is a significant challenge. Traditional toughening strategies have primarily focused on modulating the internal microstructural units of diamonds, including adjustments to stacking sequences, faults, nanotwinning, and the incorporation of amorphous phases, collectively referred to as intrinsic toughening. Here, we introduce an extrinsic toughening strategy to develop an unparalleled tough diamond composite with complex and abundant sp2-sp3 bonding interfaces, by incorporating highly dispersed multi-walled carbon nanotubes (MWCNTs) into the gaps of diamond grains to create a three-dimensional (3D) continuous MWCTNs network-toughen heterogeneous structure. The resultant composite exhibits a hardness of approximately 91.6 GPa and a fracture toughness of roughly 36.4 MPa.m1/2, which is six times higher than that of synthetic diamond and even surpasses that of tungsten alloys, surpassing the benefits achievable through intrinsic toughening alone. The remarkable toughening behavior can be attributed to the formation of numerous mixed sp2-sp3 bonding interactions at the 3D continuous network MWCNTs/diamond interfaces, which facilitate efficient energy dissipation. Our 3D continuous network heterogeneous structure design provides an effective approach for enhancing the fracture toughness of superhard materials, offering a new paradigm for the advanced composite ceramics.

cond-mat.other↗

arXiv2025

Discovery of Disease Relationships via Transcriptomic Signature Analysis Powered by Agentic AI

Modern disease classification often overlooks molecular commonalities hidden beneath divergent clinical presentations. This study introduces a transcriptomics-driven framework for discovering disease relationships by analyzing over 1300 disease-condition pairs using GenoMAS, a fully automated agentic AI system. Beyond identifying robust gene-level overlaps, we develop a novel pathway-based similarity framework that integrates multi-database enrichment analysis to quantify functional convergence across diseases. The resulting disease similarity network reveals both known comorbidities and previously undocumented cross-category links. By examining shared biological pathways, we explore potential molecular mechanisms underlying these connections-offering functional hypotheses that go beyond symptom-based taxonomies. We further show how background conditions such as obesity and hypertension modulate transcriptomic similarity, and identify therapeutic repurposing opportunities for rare diseases like autism spectrum disorder based on their molecular proximity to better-characterized conditions. In addition, this work demonstrates how biologically grounded agentic AI can scale transcriptomic analysis while enabling mechanistic interpretation across complex disease landscapes. All results are publicly accessible at github.com/KeeeeChen/Pathway_Similarity_Network.

q-bio.GN↗

arXiv2025

OpenEarthSensing: Large-Scale Fine-Grained Benchmark for Open-World Remote Sensing

The advancement of remote sensing, including satellite systems, facilitates the continuous acquisition of remote sensing imagery globally, introducing novel challenges for achieving open-world tasks. Deployed models need to continuously adjust to a constant influx of new data, which frequently exhibits diverse shifts from the data encountered during the training phase. To effectively handle the new data, models are required to detect semantic shifts, adapt to covariate shifts, and continuously update their parameters without forgetting learned knowledge, as has been considered in works on a variety of open-world tasks. However, existing studies are typically conducted within a single dataset to simulate realistic conditions, with a lack of large-scale benchmarks capable of evaluating multiple open-world tasks. In this paper, we introduce \textbf{OpenEarthSensing (OES)}, a large-scale fine-grained benchmark for open-world remote sensing. OES includes 189 scene and object categories, covering the vast majority of potential semantic shifts that may occur in the real world. Additionally, to provide a more comprehensive testbed for evaluating the generalization performance, OES encompasses five data domains with significant covariate shifts, including two RGB satellite domains, one RGB aerial domain, one multispectral RGB domain, and one infrared domain. We evaluate the baselines and existing methods for diverse tasks on OES, demonstrating that it serves as a meaningful and challenging benchmark for open-world remote sensing. The proposed dataset OES is available at https://haiv-lab.github.io/OES.

cs.CV↗

arXiv2025

RadioDUN: A Physics-Inspired Deep Unfolding Network for Radio Map Estimation

The radio map represents the spatial distribution of spectrum resources within a region, supporting efficient resource allocation and interference mitigation. However, it is difficult to construct a dense radio map as a limited number of samples can be measured in practical scenarios. While existing works have used deep learning to estimate dense radio maps from sparse samples, they are hard to integrate with the physical characteristics of the radio map. To address this challenge, we cast radio map estimation as the sparse signal recovery problem. A physical propagation model is further incorporated to decompose the problem into multiple factor optimization sub-problems, thereby reducing recovery complexity. Inspired by the existing compressive sensing methods, we propose the Radio Deep Unfolding Network (RadioDUN) to unfold the optimization process, achieving adaptive parameter adjusting and prior fitting in a learnable manner. To account for the radio propagation characteristics, we develop a dynamic reweighting module (DRM) to adaptively model the importance of each factor for the radio map. Inspired by the shadowing factor in the physical propagation model, we integrate obstacle-related factors to express the obstacle-induced signal stochastic decay. The shadowing loss is further designed to constrain the factor prediction and act as a supplementary supervised objective, which enhances the performance of RadioDUN. Extensive experiments have been conducted to demonstrate that the proposed method outperforms the state-of-the-art methods. Our code will be made publicly available upon publication.

cs.CV↗

arXiv2025

U-Net Based Healthy 3D Brain Tissue Inpainting

This paper introduces a novel approach to synthesize healthy 3D brain tissue from masked input images, specifically focusing on the task of 'ASNR-MICCAI BraTS Local Synthesis of Tissue via Inpainting'. Our proposed method employs a U-Net-based architecture, which is designed to effectively reconstruct the missing or corrupted regions of brain MRI scans. To enhance our model's generalization capabilities and robustness, we implement a comprehensive data augmentation strategy that involves randomly masking healthy images during training. Our model is trained on the BraTS-Local-Inpainting dataset and demonstrates the exceptional performance in recovering healthy brain tissue. The evaluation metrics employed, including Structural Similarity Index (SSIM), Peak Signal-to-Noise Ratio (PSNR), and Mean Squared Error (MSE), consistently yields impressive results. On the BraTS-Local-Inpainting validation set, our model achieved an SSIM score of 0.841, a PSNR score of 23.257, and an MSE score of 0.007. Notably, these evaluation metrics exhibit relatively low standard deviations, i.e., 0.103 for SSIM score, 4.213 for PSNR score and 0.007 for MSE score, which indicates that our model's reliability and consistency across various input scenarios. Our method also secured first place in the challenge.

eess.IV↗

arXiv2025

Deep Learning for Glioblastoma Morpho-pathological Features Identification: A BraTS-Pathology Challenge Solution

Glioblastoma, a highly aggressive brain tumor with diverse molecular and pathological features, poses a diagnostic challenge due to its heterogeneity. Accurate diagnosis and assessment of this heterogeneity are essential for choosing the right treatment and improving patient outcomes. Traditional methods rely on identifying specific features in tissue samples, but deep learning offers a promising approach for improved glioblastoma diagnosis. In this paper, we present our approach to the BraTS-Path Challenge 2024. We leverage a pre-trained model and fine-tune it on the BraTS-Path training dataset. Our model demonstrates poor performance on the challenging BraTS-Path validation set, as rigorously assessed by the Synapse online platform. The model achieves an accuracy of 0.392229, a recall of 0.392229, and a F1-score of 0.392229, indicating a consistent ability to correctly identify instances under the target condition. Notably, our model exhibits perfect specificity of 0.898704, showing an exceptional capacity to correctly classify negative cases. Moreover, a Matthews Correlation Coefficient (MCC) of 0.255267 is calculated, to signify a limited positive correlation between predicted and actual values and highlight our model's overall predictive power. Our solution also achieves the second place during the testing phase.

eess.IV↗

arXiv2025

Towards Human-level Intelligence via Human-like Whole-Body Manipulation

Building general-purpose intelligent robots has long been a fundamental goal of robotics. A promising approach is to mirror the evolutionary trajectory of humans: learning through continuous interaction with the environment, with early progress driven by the imitation of human behaviors. Achieving this goal presents three core challenges: (1) designing safe robotic hardware with human-level physical capabilities; (2) developing an intuitive and scalable whole-body teleoperation interface for data collection; and (3) creating algorithms capable of learning whole-body visuomotor policies from human demonstrations. To address these challenges in a unified framework, we propose Astribot Suite, a robot learning suite for whole-body manipulation aimed at general daily tasks across diverse environments. We demonstrate the effectiveness of our system on a wide range of activities that require whole-body coordination, extensive reachability, human-level dexterity, and agility. Our results show that Astribot's cohesive integration of embodiment, teleoperation interface, and learning pipeline marks a significant step towards real-world, general-purpose whole-body robotic manipulation, laying the groundwork for the next generation of intelligent robots.

cs.RO↗

arXiv2025

Amplitude mode in a multi-gap superconductor MgB$_2$ investigated by terahertz two-dimensional coherent spectroscopy

We have investigated the terahertz (THz) nonlinear response of the multigap superconductor MgB$_2$, using THz two-dimensional coherent spectroscopy (THz 2DCS). With broadband THz drive fields, we identified a nonlinear response at twice the lower superconducting gap energy $2Δ_π$ at the lowest temperatures. Using narrow-band THz driving pulses, we observed first (FH) and third harmonic responses. The FH intensity shows a monotonic increase with decreasing temperature when properly normalized by the driving field strength. This is distinct from the single-gap superconductor NbN, where the FH signal exhibited a resonant enhancement at temperatures when twice the gap energy $2Δ$ was resonant with the driving photon energy, which was interpreted to originate from the superconducting amplitude mode. Our results in MgB$_2$ are consistent with a well-defined amplitude mode only at the lowest temperatures and indicate strong damping as temperature increases. This likely indicates the importance of interband coupling in MgB$_2$ and its influence on the nature of the amplitude mode and its damping.

cond-mat.supr-con↗

arXiv2025

TableCopilot: A Table Assistant Empowered by Natural Language Conditional Table Discovery

The rise of LLM has enabled natural language-based table assistants, but existing systems assume users already have a well-formed table, neglecting the challenge of table discovery in large-scale table pools. To address this, we introduce TableCopilot, an LLM-powered assistant for interactive, precise, and personalized table discovery and analysis. We define a novel scenario, nlcTD, where users provide both a natural language condition and a query table, enabling intuitive and flexible table discovery for users of all expertise levels. To handle this, we propose Crofuma, a cross-fusion-based approach that learns and aggregates single-modal and cross-modal matching scores. Experimental results show Crofuma outperforms SOTA single-input methods by at least 12% on NDCG@5. We also release an instructional video, codebase, datasets, and other resources on GitHub to encourage community contributions. TableCopilot sets a new standard for interactive table assistants, making advanced table discovery accessible and integrated.

cs.DB↗

arXiv2025

A Comprehensive Study of Shapley Value in Data Analytics

Over the recent years, Shapley value (SV), a solution concept from cooperative game theory, has found numerous applications in data analytics (DA). This paper presents the first comprehensive study of SV used throughout the DA workflow, clarifying the key variables in defining DA-applicable SV and the essential functionalities that SV can provide for data scientists. We condense four primary challenges of using SV in DA, namely computation efficiency, approximation error, privacy preservation, and interpretability, disentangle the resolution techniques from existing arts in this field, then analyze and discuss the techniques w.r.t. each challenge and the potential conflicts between challenges.We also implement SVBench, a modular and extensible open-source framework for developing SV applications in different DA tasks, and conduct extensive evaluations to validate our analyses and discussions. Based on the qualitative and quantitative results, we identify the limitations of current efforts for applying SV to DA and highlight the directions of future research and engineering.

cs.DB↗

arXiv2025

Topology-Aware Modeling for Unsupervised Simulation-to-Reality Point Cloud Recognition

Learning semantic representations from point sets of 3D object shapes is often challenged by significant geometric variations, primarily due to differences in data acquisition methods. Typically, training data is generated using point simulators, while testing data is collected with distinct 3D sensors, leading to a simulation-to-reality (Sim2Real) domain gap that limits the generalization ability of point classifiers. Current unsupervised domain adaptation (UDA) techniques struggle with this gap, as they often lack robust, domain-insensitive descriptors capable of capturing global topological information, resulting in overfitting to the limited semantic patterns of the source domain. To address this issue, we introduce a novel Topology-Aware Modeling (TAM) framework for Sim2Real UDA on object point clouds. Our approach mitigates the domain gap by leveraging global spatial topology, characterized by low-level, high-frequency 3D structures, and by modeling the topological relations of local geometric features through a novel self-supervised learning task. Additionally, we propose an advanced self-training strategy that combines cross-domain contrastive learning with self-training, effectively reducing the impact of noisy pseudo-labels and enhancing the robustness of the adaptation process. Experimental results on three public Sim2Real benchmarks validate the effectiveness of our TAM framework, showing consistent improvements over state-of-the-art methods across all evaluated tasks. The source code of this work will be available at https://github.com/zou-longkun/TAG.git.

cs.CV↗

arXiv2025

From Generality to Mastery: Composer-Style Symbolic Music Generation via Large-Scale Pre-training

Despite progress in controllable symbolic music generation, data scarcity remains a challenge for certain control modalities. Composer-style music generation is a prime example, as only a few pieces per composer are available, limiting the modeling of both styles and fundamental music elements (e.g., melody, chord, rhythm). In this paper, we investigate how general music knowledge learned from a broad corpus can enhance the mastery of specific composer styles, with a focus on piano piece generation. Our approach follows a two-stage training paradigm. First, we pre-train a REMI-based music generation model on a large corpus of pop, folk, and classical music. Then, we fine-tune it on a small, human-verified dataset from four renowned composers, namely Bach, Mozart, Beethoven, and Chopin, using a lightweight adapter module to condition the model on style indicators. To evaluate the effectiveness of our approach, we conduct both objective and subjective evaluations on style accuracy and musicality. Experimental results demonstrate that our method outperforms ablations and baselines, achieving more precise composer-style modeling and better musical aesthetics. Additionally, we provide observations on how the model builds music concepts from the generality pre-training and refines its stylistic understanding through the mastery fine-tuning.

cs.SD↗

arXiv2025

Uniaxial stress tuning of interfacial thermal conductance in cubic BAs/4H-SiC heterostructures

Understanding interfacial thermal transport is essential for improving thermal management in high-speed power electronic devices, where the efficient removal of excess heat is a critical challenge. In this study, a machine learning interatomic potential with near first-principles accuracy was employed to investigate the interfacial thermal conductance (ITC) between [111]-oriented cubic boron arsenide (cBAs) and [0001]-oriented 4H silicon carbide (4H-SiC), as well as its dependence on uniaxial stress. Among all possible bonding configurations at the cBAs(111)/4H-SiC(0001) interface, the B-C bonded interface was identified as the most energetically favorable. Non-equilibrium molecular dynamics simulations revealed that, under ambient conditions (300 K and 0 GPa), the ITC of the B-C interface reaches 353 $\pm$ 6 MW m$^{-2}$ K$^{-1}$, and increases monotonically to 460 $\pm$ 3 MW m$^{-2}$ K$^{-1}$ under a uniaxial stress of 25 GPa perpendicular to the interface. For comparison, the As-C bonded interface exhibits a lower ITC, increasing from 233 $\pm$ 7 to 318 $\pm$ 6 MW m$^{-2}$ K$^{-1}$ over the same stress range. These results demonstrate that proper interfacial bonding and moderate uniaxial stress can significantly enhance thermal transport across the cBAs(111)/4H-SiC(0001) heterointerface, offering valuable insight for thermal design in next-generation power electronics.

cond-mat.mtrl-sci↗

arXiv2025

FLAM: Frame-Wise Language-Audio Modeling

Recent multi-modal audio-language models (ALMs) excel at text-audio retrieval but struggle with frame-wise audio understanding. Prior works use temporal-aware labels or unsupervised training to improve frame-wise capabilities, but they still lack fine-grained labeling capability to pinpoint when an event occurs. While traditional sound event detection models can precisely localize events, they are limited to pre-defined categories, making them ineffective for real-world scenarios with out-of-distribution events. In this work, we introduce FLAM, an open-vocabulary contrastive audio-language model capable of localizing specific sound events. FLAM employs a memory-efficient and calibrated frame-wise objective with logit adjustment to address spurious correlations, such as event dependencies and label imbalances during training. To enable frame-wise supervision, we leverage a large-scale dataset with diverse audio events, LLM-generated captions and simulation. Experimental results and case studies demonstrate that FLAM significantly improves the open-vocabulary localization capability while maintaining strong performance in global retrieval and downstream tasks.

cs.SD↗

arXiv2025

Point Cloud Mixture-of-Domain-Experts Model for 3D Self-supervised Learning

Point clouds, as a primary representation of 3D data, can be categorized into scene domain point clouds and object domain point clouds. Point cloud self-supervised learning (SSL) has become a mainstream paradigm for learning 3D representations. However, existing point cloud SSL primarily focuses on learning domain-specific 3D representations within a single domain, neglecting the complementary nature of cross-domain knowledge, which limits the learning of 3D representations. In this paper, we propose to learn a comprehensive Point cloud Mixture-of-Domain-Experts model (Point-MoDE) via a block-to-scene pre-training strategy. Specifically, we first propose a mixture-of-domain-expert model consisting of scene domain experts and multiple shared object domain experts. Furthermore, we propose a block-to-scene pretraining strategy, which leverages the features of point blocks in the object domain to regress their initial positions in the scene domain through object-level block mask reconstruction and scene-level block position regression. By integrating the complementary knowledge between object and scene, this strategy simultaneously facilitates the learning of both object-domain and scene-domain representations, leading to a more comprehensive 3D representation. Extensive experiments in downstream tasks demonstrate the superiority of our model.

cs.CV↗

arXiv2025

Disentangling hierarchical relaxations in glass formers via dynamic eigenmodes

Hierarchical dynamics in glass-forming systems span multiple timescales, from fast vibrations to slow structural rearrangements, appearing in both supercooled fluids and glassy states. Understanding how these diverse processes interact across timescales remains a central challenge. Here, by combining direct particle-level observations with a dynamic eigenmode approach that decomposes intermediate-timescale responses into distinct modes, we reveal the microscopic organisation of relaxation dynamics in two-dimensional colloidal systems. We identify five classes of modes characterizing hierarchical dynamics: (i) quasi-elastic modes, (ii) slow-reversible string modes contributing to dynamic heterogeneity, (iii) slow-irreversible string modes leading to flow, (iv) fast-$β$ modes with fast-reversible strings, and (v) random noise modes. The emergence of quasi-elastic modes marks the onset of glassy dynamics, while reversible string modes dominate dynamic heterogeneity throughout both supercooled and glassy regimes. Our findings offer a unified microscopic framework for understanding how distinct relaxation processes interconnect across timescales, illuminating the mechanisms driving glass formation.

cond-mat.soft↗

arXiv2025

PMA: Towards Parameter-Efficient Point Cloud Understanding via Point Mamba Adapter

Applying pre-trained models to assist point cloud understanding has recently become a mainstream paradigm in 3D perception. However, existing application strategies are straightforward, utilizing only the final output of the pre-trained model for various task heads. It neglects the rich complementary information in the intermediate layer, thereby failing to fully unlock the potential of pre-trained models. To overcome this limitation, we propose an orthogonal solution: Point Mamba Adapter (PMA), which constructs an ordered feature sequence from all layers of the pre-trained model and leverages Mamba to fuse all complementary semantics, thereby promoting comprehensive point cloud understanding. Constructing this ordered sequence is non-trivial due to the inherent isotropy of 3D space. Therefore, we further propose a geometry-constrained gate prompt generator (G2PG) shared across different layers, which applies shared geometric constraints to the output gates of the Mamba and dynamically optimizes the spatial order, thus enabling more effective integration of multi-layer information. Extensive experiments conducted on challenging point cloud datasets across various tasks demonstrate that our PMA elevates the capability for point cloud understanding to a new level by fusing diverse complementary intermediate features. Code is available at https://github.com/zyh16143998882/PMA.

cs.CV↗

arXiv2025

FloE: On-the-Fly MoE Inference on Memory-constrained GPU

With the widespread adoption of Mixture-of-Experts (MoE) models, there is a growing demand for efficient inference on memory-constrained devices. While offloading expert parameters to CPU memory and loading activated experts on demand has emerged as a potential solution, the large size of activated experts overburdens the limited PCIe bandwidth, hindering the effectiveness in latency-sensitive scenarios. To mitigate this, we propose FloE, an on-the-fly MoE inference system on memory-constrained GPUs. FloE is built on the insight that there exists substantial untapped redundancy within sparsely activated experts. It employs various compression techniques on the expert's internal parameter matrices to reduce the data movement load, combined with low-cost sparse prediction, achieving perceptible inference acceleration in wall-clock time on resource-constrained devices. Empirically, FloE achieves a 9.3x compression of parameters per expert in Mixtral-8x7B; enables deployment on a GPU with only 11GB VRAM, reducing the memory footprint by up to 8.5x; and delivers a 48.7x inference speedup compared to DeepSpeed-MII on a single GeForce RTX 3090 - all with only a 4.4$\%$ - 7.6$\%$ average performance degradation.

cs.LG↗