arXiv subjects

Haoran Chen

Publications and source records attributed to Haoran Chen.

At least 73 records · Page 4Linked to original sources

arXiv2025

Multi-Layer Visual Feature Fusion in Multimodal LLMs: Methods, Analysis, and Best Practices

Multimodal Large Language Models (MLLMs) have made significant advancements in recent years, with visual features playing an increasingly critical role in enhancing model performance. However, the integration of multi-layer visual features in MLLMs remains underexplored, particularly with regard to optimal layer selection and fusion strategies. Existing methods often rely on arbitrary design choices, leading to suboptimal outcomes. In this paper, we systematically investigate two core aspects of multi-layer visual feature fusion: (1) selecting the most effective visual layers and (2) identifying the best fusion approach with the language model. Our experiments reveal that while combining visual features from multiple stages improves generalization, incorporating additional features from the same stage typically leads to diminished performance. Furthermore, we find that direct fusion of multi-layer visual features at the input stage consistently yields superior and more stable performance across various configurations. We make all our code publicly available: https://github.com/EIT-NLP/Layer_Select_Fuse_for_MLLM.

cs.CV↗

arXiv2025

Orbital-excitation-dominated magnetization dissipation and quantum oscillation of Gilbert damping in Fe films

Using first-principles electronic structure calculation, we demonstrate the spin dissipation process in bulk Fe by orbital excitations within the energy bands of pure spin character. The variation of orbitals in the intraband transitions provides an efficient channel to convert spin to orbital angular momentum with spin-orbit interaction. This mechanism dominates the Gilbert damping of Fe below room temperature. The theoretical prediction is confirmed by the ferromagnetic resonance experiment performed on single-crystal Fe(001) films. A significant thickness-dependent damping oscillation is found at low temperature induced by the quantum well states of the corresponding energy bands. Our findings not only explain the microscopic nature of the recently reported ultralow damping of Fe-based alloys, but also help for the understanding of the transport and dissipation process of orbital currents.

cond-mat.mes-hall↗

arXiv2025

DynSegNet:Dynamic Architecture Adjustment for Adversarial Learning in Segmenting Hemorrhagic Lesions from Fundus Images

The hemorrhagic lesion segmentation plays a critical role in ophthalmic diagnosis, directly influencing early disease detection, treatment planning, and therapeutic efficacy evaluation. However, the task faces significant challenges due to lesion morphological variability, indistinct boundaries, and low contrast with background tissues. To improve diagnostic accuracy and treatment outcomes, developing advanced segmentation techniques remains imperative. This paper proposes an adversarial learning-based dynamic architecture adjustment approach that integrates hierarchical U-shaped encoder-decoder, residual blocks, attention mechanisms, and ASPP modules. By dynamically optimizing feature fusion, our method enhances segmentation performance. Experimental results demonstrate a Dice coefficient of 0.6802, IoU of 0.5602, Recall of 0.766, Precision of 0.6525, and Accuracy of 0.9955, effectively addressing the challenges in fundus image hemorrhage segmentation.[* Corresponding author.]

cs.CV↗

arXiv2025

A Benchmark for Crime Surveillance Video Analysis with Large Models

Anomaly analysis in surveillance videos is a crucial topic in computer vision. In recent years, multimodal large language models (MLLMs) have outperformed task-specific models in various domains. Although MLLMs are particularly versatile, their abilities to understand anomalous concepts and details are insufficiently studied because of the outdated benchmarks of this field not providing MLLM-style QAs and efficient algorithms to assess the model's open-ended text responses. To fill this gap, we propose a benchmark for crime surveillance video analysis with large models denoted as UCVL, including 1,829 videos and reorganized annotations from the UCF-Crime and UCF-Crime Annotation datasets. We design six types of questions and generate diverse QA pairs. Then we develop detailed instructions and use OpenAI's GPT-4o for accurate assessment. We benchmark eight prevailing MLLMs ranging from 0.5B to 40B parameters, and the results demonstrate the reliability of this bench. Moreover, we finetune LLaVA-OneVision on UCVL's training set. The improvement validates our data's high quality for video anomaly analysis.

cs.CV↗

arXiv2025

Scintillation response of Ga2O3 excited by laser accelerated ultra-high dose rate proton beam

The temporal and spectral profile of \b{eta}-Ga2O3 excited by ultra-high dose rate proton beam has been investigated. The unique short bright and broad spectra characteristics of laser-accelerated protons were utilized to investigate the scintillation response difference under different dose rate. Our results indicate that for sufficiently high dose rate delivered, the average decay time of \b{eta}-Ga2O3 decreases by a factor of two. The overlap of carriers generated by high dose rate protons enhances the nonradiative recombination like Auger recombination and exciton-exciton annihilation which shortens the decay time significantly. The study opens up new avenues for investigating the luminescent properties of other scintillator materials using laser-accelerated high dose rate proton beams.

physics.app-ph↗

arXiv2025

Anisotropic galvanomagnetic effects in single-crystal Fe(001) films elucidated by a phenomenological theory

Utilizing the phenomenological theory based on crystal symmetry operation, we have established the complete angular dependencies of the galvanomagnetic effects, encompassing both anisotropic magnetoresistance (AMR) and the planar Hall effect (PHE), for the ferromagnetic films with C4v symmetry. These dependencies were experimentally confirmed via comprehensive angular-mapping of AMR and PHE in single-crystal Fe(001) films at room temperature. We demonstrated that the intrinsic magnetization-induced effects are independent of the field strength by carefully separating the field-induced and magnetization-induced galvanomagnetic effects. Our theoretical and experimental findings highlight the absence of in-plane four-fold angular dependence in PHE, a feature prohibited by the Onsager relation in systems with C4 symmetry. This study affirms that the universal angular dependencies of AMR and PHE in single crystals can be accurately predicted by the conventional phenomenological theory.

cond-mat.mtrl-sci↗

arXiv2025

Adaptive Retention & Correction: Test-Time Training for Continual Learning

Continual learning, also known as lifelong learning or incremental learning, refers to the process by which a model learns from a stream of incoming data over time. A common problem in continual learning is the classification layer's bias towards the most recent task. Traditionally, methods have relied on incorporating data from past tasks during training to mitigate this issue. However, the recent shift in continual learning to memory-free environments has rendered these approaches infeasible. In this study, we propose a solution focused on the testing phase. We first introduce a simple Out-of-Task Detection method, OTD, designed to accurately identify samples from past tasks during testing. Leveraging OTD, we then propose: (1) an Adaptive Retention mechanism for dynamically tuning the classifier layer on past task data; (2) an Adaptive Correction mechanism for revising predictions when the model classifies data from previous tasks into classes from the current task. We name our approach Adaptive Retention & Correction (ARC). While designed for memory-free environments, ARC also proves effective in memory-based settings. Extensive experiments show that our proposed method can be plugged in to virtually any existing continual learning approach without requiring any modifications to its training procedure. Specifically, when integrated with state-of-the-art approaches, ARC achieves an average performance increase of 2.7% and 2.6% on the CIFAR-100 and Imagenet-R datasets, respectively.

cs.CV↗

arXiv2024

Interfacial Perpendicular Magnetic Anisotropy of Ultrathin Fe(001) Film Grown on CoO(001) Surface

Exploring novel systems with perpendicular magnetic anisotropy (PMA) is vital for advancing memory devices. In this study, we report an intriguing PMA system involving an ultrathin Fe layer on an antiferromagnetic (AFM) CoO(001) surface. The measured perpendicular anisotropy field is inversely proportional to the Fe thickness, indicating an interfacial origin of PMA. Temperature-dependent measurements reveal that the antiferromagnetism of CoO has a negligible effect on the PMA. By leveraging the magneto-optical Kerr effect and birefringence effect, we achieved concurrent visualization of ferromagnetic (FM) and AFM domains. A pronounced coupling effect between these domains was observed near the spin reorientation transition, contrasting sharply with areas of stronger PMA that exhibited weak coupling. This research not only establishes a new FM/AFM bilayer PMA system but also significantly advances the understanding of FM/AFM interfacial interactions.

cond-mat.mtrl-sci↗

arXiv2024

GRU-PFG: Extract Inter-Stock Correlation from Stock Factors with Graph Neural Network

The complexity of stocks and industries presents challenges for stock prediction. Currently, stock prediction models can be divided into two categories. One category, represented by GRU and ALSTM, relies solely on stock factors for prediction, with limited effectiveness. The other category, represented by HIST and TRA, incorporates not only stock factors but also industry information, industry financial reports, public sentiment, and other inputs for prediction. The second category of models can capture correlations between stocks by introducing additional information, but the extra data is difficult to standardize and generalize. Considering the current state and limitations of these two types of models, this paper proposes the GRU-PFG (Project Factors into Graph) model. This model only takes stock factors as input and extracts inter-stock correlations using graph neural networks. It achieves prediction results that not only outperform the others models relies solely on stock factors, but also achieve comparable performance to the second category models. The experimental results show that on the CSI300 dataset, the IC of GRU-PFG is 0.134, outperforming HIST's 0.131 and significantly surpassing GRU and Transformer, achieving results better than the second category models. Moreover as a model that relies solely on stock factors, it has greater potential for generalization.

q-fin.CP↗

arXiv2024

Field-free spin-orbit switching of canted magnetization in Pt/Co/Ru/RuO2(101) multilayers

Enabling field-free current-induced switching of perpendicular magnetization is essential for advancing spin-orbit-torque magnetic random access memory technology. Our research on the Pt/Co/Ru/RuO2(101) system has successfully demonstrated field-free switching through current injection along the RuO2[010] axis. We discovered that the system exhibits a tilted easy axis, inclined from the out-of-plane towards the RuO2[-101] direction. The application of current perpendicular to this tilted axis generates a substantial out-of-plane effective field, which facilitates field-free magnetization switching. Our results also indicate that adjusting the thickness of the Ru layer to optimize the tilt angle can significantly reduce the critical switching current density. This work provides a viable strategy for controlling the tilting magnetization, essential for the development of RuO2-based magnetic devices.

physics.app-ph↗

arXiv2024

To Preserve or To Compress: An In-Depth Study of Connector Selection in Multimodal Large Language Models

In recent years, multimodal large language models (MLLMs) have garnered significant attention from both industry and academia. However, there is still considerable debate on constructing MLLM architectures, particularly regarding the selection of appropriate connectors for perception tasks of varying granularities. This paper systematically investigates the impact of connectors on MLLM performance. Specifically, we classify connectors into feature-preserving and feature-compressing types. Utilizing a unified classification standard, we categorize sub-tasks from three comprehensive benchmarks, MMBench, MME, and SEED-Bench, into three task types: coarse-grained perception, fine-grained perception, and reasoning, and evaluate the performance. Our findings reveal that feature-preserving connectors excel in \emph{fine-grained perception} tasks due to their ability to retain detailed visual information. In contrast, feature-compressing connectors, while less effective in fine-grained perception tasks, offer significant speed advantages and perform comparably in \emph{coarse-grained perception} and \emph{reasoning} tasks. These insights are crucial for guiding MLLM architecture design and advancing the optimization of MLLM architectures.

cs.CL↗

arXiv2024

Magnetostatic effect on spin dynamics properties in antiferromagnetic Van der Waals material CrSBr

Van der Waals (vdW) antiferromagnets are exceptional platforms for exploring the spin dynamics of antiferromagnetic materials owing to their weak interlayer exchange coupling. In this study, we examined the antiferromagnetic resonance spectra of anisotropic Van der Waals antiferromagnet CrSBr. In addition to the ordinary resonance modes, we observed a dipolar spin wave mode when the microwave field was oriented perpendicular to the in-plane easy axis of CrSBr. Furthermore, our results uncovered a pronounced dependency of various resonant modes on the orientation of the microwave field, which is pivotal for the accurate determination of exchange coupling constants. Numerical simulations have elucidated this orientation dependence of spin dynamics arises from the magnetostatic effect. This discovery underscores the previously underappreciated significance of dipolar interactions in shaping the dynamical properties of two-dimensional AFM materials, thereby enhancing our understanding of the intrinsic dynamic properties of vdW magnets.

cond-mat.mtrl-sci↗

arXiv2024

Enhancing superconducting transition temperature of lanthanum superhydride by increasing hydrogen vacancy concentration

Various clathrate superhydride superconductors have been found to possess hydrogen deficiency in experimental samples, while their impacts on superconductivity are often neglected. In this study, we investigate the superconductivity of lanthanum superhydride with hydrogen deficiency (LaH$_{10-δ}$) from first principles using path-integral approaches. Under the effects of thermal and quantum fluctuations, hydrogen vacancies are found to diffuse within the system, leading to modifications in ion vibrations, electronic structure and electron-phonon coupling. These changes result in a non-monotonic dependence of superconducting transition temperature ($T_c$) on the vacancy concentration ($δ$). By comparing the experimental and theoretical equations of state, we suggest that $δ$ varies across samples under different pressures. This explains the positive pressure dependence of $T_c$ in experiments below 150 GPa. Remarkably, within this pressure range, we find that $T_c$ could be further raised by increasing $δ$.

cond-mat.supr-con↗

arXiv2024

A Survey on Video Diffusion Models

The recent wave of AI-generated content (AIGC) has witnessed substantial success in computer vision, with the diffusion model playing a crucial role in this achievement. Due to their impressive generative capabilities, diffusion models are gradually superseding methods based on GANs and auto-regressive Transformers, demonstrating exceptional performance not only in image generation and editing, but also in the realm of video-related research. However, existing surveys mainly focus on diffusion models in the context of image generation, with few up-to-date reviews on their application in the video domain. To address this gap, this paper presents a comprehensive review of video diffusion models in the AIGC era. Specifically, we begin with a concise introduction to the fundamentals and evolution of diffusion models. Subsequently, we present an overview of research on diffusion models in the video domain, categorizing the work into three key areas: video generation, video editing, and other video understanding tasks. We conduct a thorough review of the literature in these three key areas, including further categorization and practical contributions in the field. Finally, we discuss the challenges faced by research in this domain and outline potential future developmental trends. A comprehensive list of video diffusion models studied in this survey is available at https://github.com/ChenHsing/Awesome-Video-Diffusion-Models.

cs.CV↗

arXiv2024

Semantic Communication based on Large Language Model for Underwater Image Transmission

Underwater communication is essential for environmental monitoring, marine biology research, and underwater exploration. Traditional underwater communication faces limitations like low bandwidth, high latency, and susceptibility to noise, while semantic communication (SC) offers a promising solution by focusing on the exchange of semantics rather than symbols or bits. However, SC encounters challenges in underwater environments, including semantic information mismatch and difficulties in accurately identifying and transmitting critical information that aligns with the diverse requirements of underwater applications. To address these challenges, we propose a novel Semantic Communication (SC) framework based on Large Language Models (LLMs). Our framework leverages visual LLMs to perform semantic compression and prioritization of underwater image data according to the query from users. By identifying and encoding key semantic elements within the images, the system selectively transmits high-priority information while applying higher compression rates to less critical regions. On the receiver side, an LLM-based recovery mechanism, along with Global Vision ControlNet and Key Region ControlNet networks, aids in reconstructing the images, thereby enhancing communication efficiency and robustness. Our framework reduces the overall data size to 0.8\% of the original. Experimental results demonstrate that our method significantly outperforms existing approaches, ensuring high-quality, semantically accurate image reconstruction.

cs.CV↗

arXiv2024

PromptFusion: Decoupling Stability and Plasticity for Continual Learning

Current research on continual learning mainly focuses on relieving catastrophic forgetting, and most of their success is at the cost of limiting the performance of newly incoming tasks. Such a trade-off is referred to as the stability-plasticity dilemma and is a more general and challenging problem for continual learning. However, the inherent conflict between these two concepts makes it seemingly impossible to devise a satisfactory solution to both of them simultaneously. Therefore, we ask, "is it possible to divide them into two separate problems to conquer them independently?". To this end, we propose a prompt-tuning-based method termed PromptFusion to enable the decoupling of stability and plasticity. Specifically, PromptFusion consists of a carefully designed \stab module that deals with catastrophic forgetting and a \boo module to learn new knowledge concurrently. Furthermore, to address the computational overhead brought by the additional architecture, we propose PromptFusion-Lite which improves PromptFusion by dynamically determining whether to activate both modules for each input image. Extensive experiments show that both PromptFusion and PromptFusion-Lite achieve promising results on popular continual learning datasets for class-incremental and domain-incremental settings. Especially on Split-Imagenet-R, one of the most challenging datasets for class-incremental learning, our method can exceed state-of-the-art prompt-based methods by more than 5\% in accuracy, with PromptFusion-Lite using 14.8\% less computational resources than PromptFusion.

cs.CV↗

arXiv2024

Coexistence of Superconductivity and Superionicity in Li$_2$MgH$_{16}$

We study superconductivity in the superionic phase of the clathrate hydride Li$_2$MgH$_{16}$, where hydrogen ions diffuse among the lattice formed by lithium and magnesium ions. By employing the stochastic path-integral approach, we non-perturbatively take into account the effects of quantum diffusion and anharmonic vibrations. Our calculations reveal strong electron-ion coupling ($λ(0)$ = 3.7) and a high superconducting transition temperature ($T_c$) of 277 K under 260 GPa, at which the material is still superionic. $T_c$ is significantly suppressed compared with the result $T_c$ = 473 K obtained from the conventional approach based on the harmonic approximation. Our study, based on a first-principles approach applicable to superionic systems, indicates that the superconductivity and superionicity can coexist in Li$_2$MgH$_{16}$.

cond-mat.supr-con↗

arXiv2024

Recurrent Context Compression: Efficiently Expanding the Context Window of LLM

To extend the context length of Transformer-based large language models (LLMs) and improve comprehension capabilities, we often face limitations due to computational resources and bounded memory storage capacity. This work introduces a method called Recurrent Context Compression (RCC), designed to efficiently expand the context window length of LLMs within constrained storage space. We also investigate the issue of poor model responses when both instructions and context are compressed in downstream tasks, and propose an instruction reconstruction method to mitigate this problem. We validated the effectiveness of our approach on multiple tasks, achieving a compression rate of up to 32x on text reconstruction tasks with a BLEU4 score close to 0.95, and nearly 100\% accuracy on a passkey retrieval task with a sequence length of 1M. Finally, our method demonstrated competitive performance in long-text question-answering tasks compared to non-compressed methods, while significantly saving storage resources in long-text inference tasks. Our code, models, and demo are available at https://github.com/WUHU-G/RCC_Transformer

cs.CL↗