arXiv subjects

Yan Wang

Publications and source records attributed to Yan Wang.

At least 451 records · Page 25Linked to original sources

arXiv2025

Learning without Forgetting for Vision-Language Models

Class-Incremental Learning (CIL) or continual learning is a desired capability in the real world, which requires a learning system to adapt to new tasks without forgetting former ones. While traditional CIL methods focus on visual information to grasp core features, recent advances in Vision-Language Models (VLM) have shown promising capabilities in learning generalizable representations with the aid of textual information. However, when continually trained with new classes, VLMs often suffer from catastrophic forgetting of former knowledge. Applying VLMs to CIL poses two major challenges: 1) how to adapt the model without forgetting; and 2) how to make full use of the multi-modal information. To this end, we propose PROjectiOn Fusion (PROOF) that enables VLMs to learn without forgetting. To handle the first challenge, we propose training task-specific projections based on the frozen image/text encoders. When facing new tasks, new projections are expanded and former projections are fixed, alleviating the forgetting of old concepts. For the second challenge, we propose the fusion module to better utilize the cross-modality information. By jointly adjusting visual and textual features, the model can capture semantic information with stronger representation ability. Extensive experiments on nine benchmark datasets validate PROOF achieves state-of-the-art performance. Code is available at https://github.com/zhoudw-zdw/PROOF

cs.CV↗

arXiv2025

Semi-Supervised Vision-Centric 3D Occupancy World Model for Autonomous Driving

Understanding world dynamics is crucial for planning in autonomous driving. Recent methods attempt to achieve this by learning a 3D occupancy world model that forecasts future surrounding scenes based on current observation. However, 3D occupancy labels are still required to produce promising results. Considering the high annotation cost for 3D outdoor scenes, we propose a semi-supervised vision-centric 3D occupancy world model, PreWorld, to leverage the potential of 2D labels through a novel two-stage training paradigm: the self-supervised pre-training stage and the fully-supervised fine-tuning stage. Specifically, during the pre-training stage, we utilize an attribute projection head to generate different attribute fields of a scene (e.g., RGB, density, semantic), thus enabling temporal supervision from 2D labels via volume rendering techniques. Furthermore, we introduce a simple yet effective state-conditioned forecasting module to recursively forecast future occupancy and ego trajectory in a direct manner. Extensive experiments on the nuScenes dataset validate the effectiveness and scalability of our method, and demonstrate that PreWorld achieves competitive performance across 3D occupancy prediction, 4D occupancy forecasting and motion planning tasks.

cs.CV↗

arXiv2025

CAMSIC: Content-aware Masked Image Modeling Transformer for Stereo Image Compression

Existing learning-based stereo image codec adopt sophisticated transformation with simple entropy models derived from single image codecs to encode latent representations. However, those entropy models struggle to effectively capture the spatial-disparity characteristics inherent in stereo images, which leads to suboptimal rate-distortion results. In this paper, we propose a stereo image compression framework, named CAMSIC. CAMSIC independently transforms each image to latent representation and employs a powerful decoder-free Transformer entropy model to capture both spatial and disparity dependencies, by introducing a novel content-aware masked image modeling (MIM) technique. Our content-aware MIM facilitates efficient bidirectional interaction between prior information and estimated tokens, which naturally obviates the need for an extra Transformer decoder. Experiments show that our stereo image codec achieves state-of-the-art rate-distortion performance on two stereo image datasets Cityscapes and InStereo2K with fast encoding and decoding speed. Code is available at https://github.com/Xinjie-Q/CAMSIC.

eess.IV↗

arXiv2025

A Comprehensive Review of Protein Language Models

At the intersection of the rapidly growing biological data landscape and advancements in Natural Language Processing (NLP), protein language models (PLMs) have emerged as a transformative force in modern research. These models have achieved remarkable progress, highlighting the need for timely and comprehensive overviews. However, much of the existing literature focuses narrowly on specific domains, often missing a broader analysis of PLMs. This study provides a systematic review of PLMs from a macro perspective, covering key historical milestones and current mainstream trends. We focus on the models themselves and their evaluation metrics, exploring aspects such as model architectures, positional encoding, scaling laws, and datasets. In the evaluation section, we discuss benchmarks and downstream applications. To further support ongoing research, we introduce relevant mainstream tools. Lastly, we critically examine the key challenges and limitations in this rapidly evolving field.

q-bio.BM↗

arXiv2025

Effects of orbital eccentricity on continuous gravitational waveforms from triaxially deformed precessing neutron stars in tight binaries

The successful detection of continuous gravitational waves from spinning neutron stars (NSs) will shape our understanding of the physical properties of dense matter under extreme conditions. Binary population synthesis simulations show that forthcoming space-borne gravitational wave detectors may be capable of detecting some tight Galactic double NSs with 10-min orbital periods. Successfully searching for continuous waves from the individual NS in such a close binary demands extremely precise waveform templates considering the interaction between the NS and its companion. Unlike the isolated formation channel, double NS systems from the dynamical formation channel have moderate to high orbital eccentricities. To accommodate these systems, we generalize the analytical waveforms from triaxial nonaligned NSs under spin-orbit coupling derived by Feng \textit{et al.} [\href{https://journals.aps.org/prd/abstract/10.1103/PhysRevD.108.063035}{Phys. Rev. D 108, 063035 (2023)}] to incorporate the effects of the orbital eccentricity. Our findings suggest that for binaries formed through isolated binary evolution, the impact of eccentricity on the continuous waves of their NSs can be neglected. In contrast, for those formed through dynamical processes, it is necessary to consider eccentricity, as high-eccentricity orbits can result in a fitting factor of $\lesssim 0.97$ (0.9) within approximately 0.5 (1) to 2 (5) yr of a coherent search (at wave frequencies of 100 and 200 Hz). Once the continuous waves from spinning NSs in tight binaries are detected, the relative measurement accuracy of eccentricity can reach $Δe / e \sim O(10^{-7})$ for a signal-to-noise ratio of $O(100)$ based on the Fisher information matrix, bearing significant implications for understanding the formation mechanisms of double NS systems.

gr-qc↗

arXiv2025

Hierarchical search method for gravitational waves from stellar-mass binary black holes in noisy space-based detector data

Future space-based laser interferometric detectors, such as LISA, will be able to detect gravitational waves (GWs) generated during the inspiral phase of stellar-mass binary black holes (SmBBHs). The detection and characterization of GWs from SmBBHs poses a formidable data analysis challenge, arising from the large number of wave cycles that make the search extremely sensitive to mismatches in signal and template parameters in a likelihood-based approach. This makes the search for the maximum of the likelihood function over the signal parameter space an extremely difficult task. We present a data analysis method that addresses this problem using both algorithmic innovations and hardware acceleration driven by GPUs. The method follows a hierarchical approach in which a semi-coherent $\mathcal{F}$-statistic is computed with different numbers of frequency domain partitions at different stages, with multiple particle swarm optimization (PSO) runs used in each stage for global optimization. An important step in the method is the judicious partitioning of the parameter space at each stage to improve the convergence probability of PSO and avoid premature convergence to noise-induced secondary maxima. The hierarchy of stages confines the semi-coherent searches to progressively smaller parameter ranges, with the final stage performing a search for the global maximum of the fully-coherent $\mathcal{F}$-statistic. We test our method on 2.5 years of a single LISA TDI combination and find that for an injected SmBBH signal with a SNR between $\approx 11$ and $\approx 14$, the method can estimate (i) the chirp mass with a relative error of $\lesssim 0.01\%$, (ii) the time of coalescence within $\approx 100$ sec, (iii) the sky location within $\approx 0.2$ ${\rm deg}^2$, and (iv) orbital eccentricity at a fiducial signal frequency of 10 mHz with a relative error of $\lesssim 1\%$. (abr.)

astro-ph.IM↗

arXiv2025

Idempotence and Perceptual Image Compression

Idempotence is the stability of image codec to re-compression. At the first glance, it is unrelated to perceptual image compression. However, we find that theoretically: 1) Conditional generative model-based perceptual codec satisfies idempotence; 2) Unconditional generative model with idempotence constraint is equivalent to conditional generative codec. Based on this newfound equivalence, we propose a new paradigm of perceptual image codec by inverting unconditional generative model with idempotence constraints. Our codec is theoretically equivalent to conditional generative codec, and it does not require training new models. Instead, it only requires a pre-trained mean-square-error codec and unconditional generative model. Empirically, we show that our proposed approach outperforms state-of-the-art methods such as HiFiC and ILLM, in terms of Fréchet Inception Distance (FID). The source code is provided in https://github.com/tongdaxu/Idempotence-and-Perceptual-Image-Compression.

eess.IV↗

arXiv2025

IROAM: Improving Roadside Monocular 3D Object Detection Learning from Autonomous Vehicle Data Domain

In autonomous driving, The perception capabilities of the ego-vehicle can be improved with roadside sensors, which can provide a holistic view of the environment. However, existing monocular detection methods designed for vehicle cameras are not suitable for roadside cameras due to viewpoint domain gaps. To bridge this gap and Improve ROAdside Monocular 3D object detection, we propose IROAM, a semantic-geometry decoupled contrastive learning framework, which takes vehicle-side and roadside data as input simultaneously. IROAM has two significant modules. In-Domain Query Interaction module utilizes a transformer to learn content and depth information for each domain and outputs object queries. Cross-Domain Query Enhancement To learn better feature representations from two domains, Cross-Domain Query Enhancement decouples queries into semantic and geometry parts and only the former is used for contrastive learning. Experiments demonstrate the effectiveness of IROAM in improving roadside detector's performance. The results validate that IROAM has the capabilities to learn cross-domain information.

cs.CV↗

arXiv2025

AccidentBlip: Agent of Accident Warning based on MA-former

In complex transportation systems, accurately sensing the surrounding environment and predicting the risk of potential accidents is crucial. Most existing accident prediction methods are based on temporal neural networks, such as RNN and LSTM. Recent multimodal fusion approaches improve vehicle localization through 3D target detection and assess potential risks by calculating inter-vehicle distances. However, these temporal networks and multimodal fusion methods suffer from limited detection robustness and high economic costs. To address these challenges, we propose AccidentBlip, a vision-only framework that employs our self-designed Motion Accident Transformer (MA-former) to process each frame of video. Unlike conventional self-attention mechanisms, MA-former replaces Q-former's self-attention with temporal attention, allowing the query corresponding to the previous frame to generate the query input for the next frame. Additionally, we introduce a residual module connection between queries of consecutive frames to enhance the model's temporal processing capabilities. For complex V2V and V2X scenarios, AccidentBlip adapts by concatenating queries from multiple cameras, effectively capturing spatial and temporal relationships. In particular, AccidentBlip achieves SOTA performance in both accident detection and prediction tasks on the DeepAccident dataset. It also outperforms current SOTA methods in V2V and V2X scenarios, demonstrating a superior capability to understand complex real-world environments.

cs.AI↗

arXiv2025

Role of interface mixing on coherent heat conduction in periodic and aperiodic superlattices

Superlattices (SLs) can induce phonon coherence through the periodic layering of two or more materials, enabling tailored thermal transport properties. While most theoretical studies assume atomically sharp, perfect interfaces, real SLs often feature atomic interdiffusion spanning approximately a single atomic layer or more. Such interface mixing can significantly influence phonon coherence and transport behavior. In this study, we employ atomistic wave-packet simulations to systematically investigate the effects of interface mixing on coherent heat conduction. Our analysis identifies two competing mechanisms that govern phonon transport across mixed interfaces: (1) Interface mixing disrupts coherent mode-conversion effects arising from the interface arrangement. (2) The disorder enhances the potential for interference events, generating additional coherent phonon transport pathways. The second mechanism enhances the transmission of Bragg-reflected modes in periodic SLs and most phonons in aperiodic SLs, which otherwise lack coherent mode-conversion in perfect structures. Conversely, the first mechanism dominates in periodic SLs for non-Bragg-reflected modes, where transmission is already high due to substantial mode-conversion. These findings provide insights into the interplay between interface imperfections and phonon coherence.

cond-mat.mtrl-sci↗

arXiv2025

Wafer-scale Integration of Single-Crystalline MoS$_2$ for Flexible Electronics Enabled by Oxide Dry-transfer

Atomically thin, single-crystalline transition metal dichalcogenides (TMDCs) grown via chemical vapor deposition (CVD) on sapphire substrates exhibit exceptional mechanical and electrical properties, positioning them as excellent channel materials for flexible electronics. However, conventional wet-transfer processes for integrating these materials onto flexible substrates often introduce surface contamination, significantly degrading device performance. Here, we present a wafer-scale dry-transfer technique using a high-dielectric oxide as the transfer medium, enabling the integration of 4-inch single-crystalline MoS$_2$ onto flexible substrates. This method eliminates contact with polymers or solvents, thus preserving the intrinsic electronic properties of MoS$_2$. As a result, the fabricated flexible field-effect transistor (FET) arrays exhibit remarkable performance, with a mobility of 117 cm$^2$/Vs, a subthreshold swing of 68.8 mV dec$^{-1}$, and an ultra-high current on/off ratio of $10^{12}$-values comparable to those achieved on rigid substrates. Leveraging the outstanding electrical characteristics, we demonstrated MoS$_2$-based flexible inverters operating in the subthreshold regime, achieving both a high gain of 218 and ultra-low power consumption of 1.4 pW/$μ$m. Additionally, we integrated a flexible tactile sensing system driven by active-matrix MoS$_2$ FET arrays onto a robotic gripper, enabling real-time object identification. These findings demonstrate the simultaneous achievement of high electrical performance and flexibility, highlighting the immense potential of single-crystalline TMDC-based flexible electronics for real-world applications.

physics.app-ph↗

arXiv2025

Rainbow perfect matchings in 3-partite 3-uniform hypergraphs

Let $m,n,r,s$ be nonnegative integers such that $n\ge m=3r+s$ and $1\leq s\leq 3$. Let \[δ(n,r,s)=\left\{\begin{array}{ll} n^2-(n-r)^2 &\text{if}\ s=1 , \\[5pt] n^2-(n-r+1)(n-r-1) &\text{if}\ s=2,\\[5pt] n^2 - (n-r)(n-r-1) &\text{if}\ s=3. \end{array}\right.\] We show that there exists a constant $n_0 > 0$ such that if $F_1,\ldots, F_n$ are 3-partite 3-graphs with $n\ge n_0$ vertices in each partition class and minimum vertex degree of $F_i$ is at least $δ(n,r,s)+1$ for $i \in [n]$ then $\{F_1,\ldots,F_n\}$ admits a rainbow perfect matching. This generalizes a result of Lo and Markström on the vertex degree threshold for the existence of perfect matchings in 3-partite 3-graphs. In this proof, we use a fractional rainbow matching theory obtained by Aharoni et al. to find edge-disjoint fractional perfect matching.

math.CO↗

arXiv2025

The Oscars of AI Theater: A Survey on Role-Playing with Language Models

This survey explores the burgeoning field of role-playing with language models, focusing on their development from early persona-based models to advanced character-driven simulations facilitated by Large Language Models (LLMs). Initially confined to simple persona consistency due to limited model capabilities, role-playing tasks have now expanded to embrace complex character portrayals involving character consistency, behavioral alignment, and overall attractiveness. We provide a comprehensive taxonomy of the critical components in designing these systems, including data, models and alignment, agent architecture and evaluation. This survey not only outlines the current methodologies and challenges, such as managing dynamic personal profiles and achieving high-level persona consistency but also suggests avenues for future research in improving the depth and realism of role-playing applications. The goal is to guide future research by offering a structured overview of current methodologies and identifying potential areas for improvement. Related resources and papers are available at https://github.com/nuochenpku/Awesome-Role-Play-Papers.

cs.AI↗

arXiv2025

A Holistically Point-guided Text Framework for Weakly-Supervised Camouflaged Object Detection

Weakly-Supervised Camouflaged Object Detection (WSCOD) has gained popularity for its promise to train models with weak labels to segment objects that visually blend into their surroundings. Recently, some methods using sparsely-annotated supervision shown promising results through scribbling in WSCOD, while point-text supervision remains underexplored. Hence, this paper introduces a novel holistically point-guided text framework for WSCOD by decomposing into three phases: segment, choose, train. Specifically, we propose Point-guided Candidate Generation (PCG), where the point's foreground serves as a correction for the text path to explicitly correct and rejuvenate the loss detection object during the mask generation process (SEGMENT). We also introduce a Qualified Candidate Discriminator (QCD) to choose the optimal mask from a given text prompt using CLIP (CHOOSE), and employ the chosen pseudo mask for training with a self-supervised Vision Transformer (TRAIN). Additionally, we developed a new point-supervised dataset (P2C-COD) and a text-supervised dataset (T-COD). Comprehensive experiments on four benchmark datasets demonstrate our method outperforms state-of-the-art methods by a large margin, and also outperforms some existing fully-supervised camouflaged object detection methods.

cs.CV↗

arXiv2025

Magnetic topological Weyl fermions in half-metallic In$_2$CoSe$_4$

Magnetic Weyl semimetals (WSM) have recently attracted much attention due to their potential in realizing strong anomalous Hall effects. Yet, how to design such systems remains unclear. Based on first-principles calculations, we show here that the ferromagnetic half-metallic compound In$_2$CoSe$_4$ has several pairs of Weyl points and is hence a good candidate for magnetic WSM. These Weyl points would approach the Fermi level gradually as the Hubbard $U$ increases, and finally disappear after a critical value $U_c$. The range of the Hubbard $U$ that can realize the magnetic WSM state can be expanded by pressure, manifesting the practical utility of the present prediction. Moreover, by generating two surface terminations at Co or In atom after cleaving the compound at the Co-Se bonds, the nontrivial Fermi arcs connecting one pair of Weyl points with opposite chirality are discovered in surface states. Furthermore, it is possible to observe the nontrivial surface state experimentally, e.g., angle-resolved photoemission spectroscopy (ARPES) measurements. As such, the present findings imply strongly a new magnetic WSM which may host a large anomalous Hall conductivity.

cond-mat.mtrl-sci↗

arXiv2025

BASIC: Semi-supervised Multi-organ Segmentation with Balanced Subclass Regularization and Semantic-conflict Penalty

Semi-supervised learning (SSL) has shown notable potential in relieving the heavy demand of dense prediction tasks on large-scale well-annotated datasets, especially for the challenging multi-organ segmentation (MoS). However, the prevailing class-imbalance problem in MoS caused by the substantial variations in organ size exacerbates the learning difficulty of the SSL network. To address this issue, in this paper, we propose an innovative semi-supervised network with BAlanced Subclass regularIzation and semantic-Conflict penalty mechanism (BASIC) to effectively learn the unbiased knowledge for semi-supervised MoS. Concretely, we construct a novel auxiliary subclass segmentation (SCS) task based on priorly generated balanced subclasses, thus deeply excavating the unbiased information for the main MoS task with the fashion of multi-task learning. Additionally, based on a mean teacher framework, we elaborately design a balanced subclass regularization to utilize the teacher predictions of SCS task to supervise the student predictions of MoS task, thus effectively transferring unbiased knowledge to the MoS subnetwork and alleviating the influence of the class-imbalance problem. Considering the similar semantic information inside the subclasses and their corresponding original classes (i.e., parent classes), we devise a semantic-conflict penalty mechanism to give heavier punishments to the conflicting SCS predictions with wrong parent classes and provide a more accurate constraint to the MoS predictions. Extensive experiments conducted on two publicly available datasets, i.e., the WORD dataset and the MICCAI FLARE 2022 dataset, have verified the superior performance of our proposed BASIC compared to other state-of-the-art methods.

cs.CV↗

arXiv2025

P3S-Diffusion:A Selective Subject-driven Generation Framework via Point Supervision

Recent research in subject-driven generation increasingly emphasizes the importance of selective subject features. Nevertheless, accurately selecting the content in a given reference image still poses challenges, especially when selecting the similar subjects in an image (e.g., two different dogs). Some methods attempt to use text prompts or pixel masks to isolate specific elements. However, text prompts often fall short in precisely describing specific content, and pixel masks are often expensive. To address this, we introduce P3S-Diffusion, a novel architecture designed for context-selected subject-driven generation via point supervision. P3S-Diffusion leverages minimal cost label (e.g., points) to generate subject-driven images. During fine-tuning, it can generate an expanded base mask from these points, obviating the need for additional segmentation models. The mask is employed for inpainting and aligning with subject representation. The P3S-Diffusion preserves fine features of the subjects through Multi-layers Condition Injection. Enhanced by the Attention Consistency Loss for improved training, extensive experiments demonstrate its excellent feature preservation and image generation capabilities.

cs.CV↗

arXiv2025

DreamDrive: Generative 4D Scene Modeling from Street View Images

Synthesizing photo-realistic visual observations from an ego vehicle's driving trajectory is a critical step towards scalable training of self-driving models. Reconstruction-based methods create 3D scenes from driving logs and synthesize geometry-consistent driving videos through neural rendering, but their dependence on costly object annotations limits their ability to generalize to in-the-wild driving scenarios. On the other hand, generative models can synthesize action-conditioned driving videos in a more generalizable way but often struggle with maintaining 3D visual consistency. In this paper, we present DreamDrive, a 4D spatial-temporal scene generation approach that combines the merits of generation and reconstruction, to synthesize generalizable 4D driving scenes and dynamic driving videos with 3D consistency. Specifically, we leverage the generative power of video diffusion models to synthesize a sequence of visual references and further elevate them to 4D with a novel hybrid Gaussian representation. Given a driving trajectory, we then render 3D-consistent driving videos via Gaussian splatting. The use of generative priors allows our method to produce high-quality 4D scenes from in-the-wild driving data, while neural rendering ensures 3D-consistent video generation from the 4D scenes. Extensive experiments on nuScenes and street view images demonstrate that DreamDrive can generate controllable and generalizable 4D driving scenes, synthesize novel views of driving videos with high fidelity and 3D consistency, decompose static and dynamic elements in a self-supervised manner, and enhance perception and planning tasks for autonomous driving.

cs.CV↗