arXiv subjects

Jun Zhang

Publications and source records attributed to Jun Zhang.

At least 451 records · Page 25Linked to original sources

arXiv2024

FairAdapter: Detecting AI-generated Images with Improved Fairness

The high-quality, realistic images generated by generative models pose significant challenges for exposing them.So far, data-driven deep neural networks have been justified as the most efficient forensics tools for the challenges. However, they may be over-fitted to certain semantics, resulting in considerable inconsistency in detection performance across different contents of generated samples. It could be regarded as an issue of detection fairness. In this paper, we propose a novel framework named Fairadapter to tackle the issue. In comparison with existing state-of-the-art methods, our model achieves improved fairness performance. Our project: https://github.com/AppleDogDog/FairnessDetection

cs.CV↗

arXiv2024

Advancing Pose-Guided Image Synthesis with Progressive Conditional Diffusion Models

Recent work has showcased the significant potential of diffusion models in pose-guided person image synthesis. However, owing to the inconsistency in pose between the source and target images, synthesizing an image with a distinct pose, relying exclusively on the source image and target pose information, remains a formidable challenge. This paper presents Progressive Conditional Diffusion Models (PCDMs) that incrementally bridge the gap between person images under the target and source poses through three stages. Specifically, in the first stage, we design a simple prior conditional diffusion model that predicts the global features of the target image by mining the global alignment relationship between pose coordinates and image appearance. Then, the second stage establishes a dense correspondence between the source and target images using the global features from the previous stage, and an inpainting conditional diffusion model is proposed to further align and enhance the contextual features, generating a coarse-grained person image. In the third stage, we propose a refining conditional diffusion model to utilize the coarsely generated image from the previous stage as a condition, achieving texture restoration and enhancing fine-detail consistency. The three-stage PCDMs work progressively to generate the final high-quality and high-fidelity synthesized image. Both qualitative and quantitative results demonstrate the consistency and photorealism of our proposed PCDMs under challenging scenarios.The code and model will be available at https://github.com/tencent-ailab/PCDMs.

cs.CV↗

arXiv2024

Towards 250-m gigabits-per-second underwater wireless optical communication using a low-complexity ANN equalizer

The breakthroughs of communication distance and data rate have been eagerly anticipated by scientists in the area of underwater wireless optical communication (UWOC), which is seriously limited by the obvious aquatic attenuation in underwater channel. High-power laser source and ultra-sensitive photodetector are straightforward to extend the UWOC distance. However, nonlinear impairments caused by bandwidth-limited high-power transmitter and sensitive receiver severely degrade the data rate of long-distance UWOC. In this paper, we develop a UWOC system using a high-power transmitter by beam combining of 8-channel cascaded laser diodes (LD) and a sensitive receiver by a silicon photomultiplier (SiPM). The combined linear equalizer and low-complexity Artificial Neural Network (ANN) equalizer are used to achieve 1-Gbps data transmission over a 250-m UWOC system. To the best of our knowledge, this is the first Gbps-level UWOC experimental demonstration in >250-meter underwater transmission that has ever been reported. To lower the complexity of the ANN equalizer, a linear equalizer is applied first in order to prune the input size of the ANN equalizer. The optimal input size of the ANN equalizer is identified as 9. And the ANN architecture consists of two hidden layers, with 10 neurons in the first layer and a single neuron in the second layer. The performance of the proposed ANN-based system is compared with that of systems employing Volterra and linear equalizers. The bit error rate (BER) at data rate of 1 Gbps over a 250-m UWOC is reduced to with the combined linear and ANN equalizer, which is below the hard-decision forward error correction (HD-FEC) limit. In contrast, the linear and Volterra equalizer-based systems achieve data rates of 500 Mbps and 750 Mbps, respectively.

physics.optics↗

arXiv2024

A Dense Subframe-based SLAM Framework with Side-scan Sonar

Side-scan sonar (SSS) is a lightweight acoustic sensor that is commonly deployed on autonomous underwater vehicles (AUVs) to provide high-resolution seafloor images. However, leveraging side-scan images for simultaneous localization and mapping (SLAM) presents a notable challenge, primarily due to the difficulty of establishing sufficient amount of accurate correspondences between these images. To address this, we introduce a novel subframe-based dense SLAM framework utilizing side-scan sonar data, enabling effective dense matching in overlapping regions of paired side-scan images. With each image being evenly divided into subframes, we propose a robust estimation pipeline to estimate the relative pose between each paired subframes, by using a good inlier set identified from dense correspondences. These relative poses are then integrated as edge constraints in a factor graph to optimize the AUV pose trajectory. The proposed framework is evaluated on three real datasets collected by a Hugin AUV. Among one of them includes manually-annotated keypoint correspondences as ground truth and is used for evaluation of pose trajectory. We also present a feasible way of evaluating mapping quality against multi-beam echosounder (MBES) data without the influence of pose. Experimental results demonstrate that our approach effectively mitigates drift from the dead-reckoning (DR) system and enables quasi-dense bathymetry reconstruction. An open-source implementation of this work is available.

cs.RO↗

arXiv2024

Exciton Enhanced Giant Correlated Stoke AntiStokes Scattering of Multiorder Phonons in Semiconductor

The correlated Stoke antiStokes (SaS) scattering plays a crucial role in quantum information processing, such as heralded light sources, Fock state dynamics, and write read protocol for quantum memory. However, several reported materials exhibit low degree of SaS correlation and require high-power pulse laser excitation, limiting further applications. Herein, we explore the giant correlated multiorder SaS scattering under low power continuous laser excitation through red-sideband resonance of exciton in semiconductor ZnTe nanobelts. At low temperatures, we observe an unexpectedly strong anti-Stokes signal for multiorder longitudinal optical phonons, with SaS correlations two or four orders of magnitude larger than reported results. Furthermore, we observed the mitigation of laser heating effect for longitudinal optical phonon in SaS scattering. This finding paves a new pathway to study multiorder quantum correlated photon pairs produced through exciton-resonant Raman scattering.

cond-mat.mes-hall↗

arXiv2024

Cavity-enhanced circular dichroism in a van der Waals antiferromagnet

Broken symmetry plays a pivotal role in determining the macroscopic electrical, optical, magnetic, and topological properties of materials. Circular dichroism (CD) has been widely employed to probe broken symmetry in various systems, from small molecules to bulk crystals, but designing CD responses on demand remains a challenge, especially for antiferromagnetic materials. Here, we develop a cavity-enhanced CD technique to sensitively probe the magnetic order and broken symmetry in the van der Waals antiferromagnet FePS3. By introducing interfacial inversion asymmetry in cavity-coupled FePS3 crystals, we demonstrate that the induced CD is strongly coupled with the zig-zag antiferromagnetic order of FePS3 and can be tuned both spectrally and in magnitude by varying the cavity length and FePS3 thickness. Our findings open new avenues for using cavity-modulated CD as a sensitive diagnostic probe to detect weak broken symmetries, particularly at hidden interfaces, and in systems exhibiting hidden spin polarization or strong correlations.

cond-mat.mes-hall↗

arXiv2024

DimensionX: Create Any 3D and 4D Scenes from a Single Image with Controllable Video Diffusion

In this paper, we introduce \textbf{DimensionX}, a framework designed to generate photorealistic 3D and 4D scenes from just a single image with video diffusion. Our approach begins with the insight that both the spatial structure of a 3D scene and the temporal evolution of a 4D scene can be effectively represented through sequences of video frames. While recent video diffusion models have shown remarkable success in producing vivid visuals, they face limitations in directly recovering 3D/4D scenes due to limited spatial and temporal controllability during generation. To overcome this, we propose ST-Director, which decouples spatial and temporal factors in video diffusion by learning dimension-aware LoRAs from dimension-variant data. This controllable video diffusion approach enables precise manipulation of spatial structure and temporal dynamics, allowing us to reconstruct both 3D and 4D representations from sequential frames with the combination of spatial and temporal dimensions. Additionally, to bridge the gap between generated videos and real-world scenes, we introduce a trajectory-aware mechanism for 3D generation and an identity-preserving denoising strategy for 4D generation. Extensive experiments on various real-world and synthetic datasets demonstrate that DimensionX achieves superior results in controllable video generation, as well as in 3D and 4D scene generation, compared with previous methods.

cs.CV↗

arXiv2024

Vista: A Generalizable Driving World Model with High Fidelity and Versatile Controllability

World models can foresee the outcomes of different actions, which is of paramount importance for autonomous driving. Nevertheless, existing driving world models still have limitations in generalization to unseen environments, prediction fidelity of critical details, and action controllability for flexible application. In this paper, we present Vista, a generalizable driving world model with high fidelity and versatile controllability. Based on a systematic diagnosis of existing methods, we introduce several key ingredients to address these limitations. To accurately predict real-world dynamics at high resolution, we propose two novel losses to promote the learning of moving instances and structural information. We also devise an effective latent replacement approach to inject historical frames as priors for coherent long-horizon rollouts. For action controllability, we incorporate a versatile set of controls from high-level intentions (command, goal point) to low-level maneuvers (trajectory, angle, and speed) through an efficient learning strategy. After large-scale training, the capabilities of Vista can seamlessly generalize to different scenarios. Extensive experiments on multiple datasets show that Vista outperforms the most advanced general-purpose video generator in over 70% of comparisons and surpasses the best-performing driving world model by 55% in FID and 27% in FVD. Moreover, for the first time, we utilize the capacity of Vista itself to establish a generalizable reward for real-world action evaluation without accessing the ground truth actions.

cs.CV↗

arXiv2024

Quantum Interference and Optical Tuning of Self-Trapped Exciton State in Double Halide Perovskite

Self-trapped excitons (STEs), renowned for their unique radiative properties, have been harnessed in diverse photonic devices. Yet, a full comprehension and manipulation of STEs remain elusive. In this study, we present novel experimental and theoretical evidence of the hybrid nature and optical tuning of the STEs state in Cs2Ag0.4Na0.6InCl6. The detection of Fano resonance in the laser energy-dependent Raman and photoluminescence spectra indicates the emergence of an exciton-phonon hybrid state, a result of the robust quantum interference between the discrete phonon and continuous exciton states. Moreover, we showcase the ability to continuously adjust this hybrid state with the energy and intensity of the laser field. These significant findings lay the foundation for a comprehensive understanding of the nature of STE and its potential for state control.

cond-mat.mtrl-sci↗

arXiv2024

Bidirectional Stereo Image Compression with Cross-Dimensional Entropy Model

With the rapid advancement of stereo vision technologies, stereo image compression has emerged as a crucial field that continues to draw significant attention. Previous approaches have primarily employed a unidirectional paradigm, where the compression of one view is dependent on the other, resulting in imbalanced compression. To address this issue, we introduce a symmetric bidirectional stereo image compression architecture, named BiSIC. Specifically, we propose a 3D convolution based codec backbone to capture local features and incorporate bidirectional attention blocks to exploit global features. Moreover, we design a novel cross-dimensional entropy model that integrates various conditioning factors, including the spatial context, channel context, and stereo dependency, to effectively estimate the distribution of latent representations for entropy coding. Extensive experiments demonstrate that our proposed BiSIC outperforms conventional image/video compression standards, as well as state-of-the-art learning-based methods, in terms of both PSNR and MS-SSIM.

eess.IV↗

arXiv2024

Influences of ${Z=100}$ and ${N=152}$ deformed shells on ${ K^π=8^{-} }$ isomers and rotational bands in ${N = 150}$ isotones

The $K^π=8^{-}$ isomeric states and rotational bands in the even-even $N = 150$ isotones with $94 \leqslant Z \leqslant 104$ are investigated by the cranked shell model (CSM) with pairing correlations treated by the particle-number-conserving (PNC) method. The experimental bandhead energies and kinematic moments of inertia (MOIs) are reproduced quite well by the PNC-CSM calculation. The two-neutron state with configuration $ν9/2^{-}[734] \otimes ν7/2^{+}[624]$ is the lowest $8^{-}$ state for these isomers. This is a demonstration of the deformed neutron shell at $N=152$. Low-lying two proton $π^{2}8^{-}$($π9/2^{+}[624] \otimes π7/2^{-}[514]$) configuration state is predicted only for $^{252}$No and $^{254}$Rf due to the deformed proton shell at $Z=100$. A distinct upbending is observed for the $ν^{2}8^{-}$ bands in the lighter isotones while it is absent for bands in the heavier ones. The upbending of the $ν^{2}8^{-}$ band at frequency $\hbarω\approx 0.20$ MeV in $^{244}$Pu attributes to the sudden proton alignment of the interference term $j_x(\pi5/2^{+}[642]\otimes\pi7/2^{+}[633])$. The irregularity of MOI observed in the $K^π=8^{-}$ band of $^{252}$No can be explained by the mixing of the $ν^{2}8^{-}$($ν9/2^{-}[734] \otimes ν7/2^{+}[624]$) and $π^{2}8^{-}$($π9/2^{+}[624] \otimes π7/2^{-}[514]$) configurations. The $20\%-30\%$ increase of the bandhead $J^{(1)}$ for the $8^{-}$ bands comparing to the ground-state band is attributed to the $\sim 5\%$ pairing gap reduction of the two-neutron $ν7/2^{+}[624] \otimes ν9/2^{-}[734]$ configuration state comparing to the ground-state band.

nucl-th↗

arXiv2024

Position-Aided Semantic Communication for Efficient Image Transmission: Design, Implementation, and Experimental Results

Semantic communication, augmented by knowledge bases (KBs), offers substantial reductions in transmission overhead and resilience to errors. However, existing methods predominantly rely on end-to-end training to construct KBs, often failing to fully capitalize on the rich information available at communication devices. Motivated by the growing convergence of sensing and communication, we introduce a novel Position-Aided Semantic Communication (PASC) framework, which integrates localization into semantic transmission. This framework is particularly designed for position-based image communication, such as real-time uploading of outdoor camera-view images. By utilizing the position, the framework retrieves corresponding maps, and then an advanced foundation model (FM)-driven view generator is employed to synthesize images closely resembling the target images. The PASC framework further leverages the FM to fuse the synthesized image with deviations from the real one, enhancing semantic reconstruction. Notably, the framework is highly flexible, capable of adapting to dynamic content and fluctuating channel conditions through a novel FM-based parameter optimization strategy. Additionally, the challenges of real-time deployment are addressed, with the development of a hardware testbed to validate the framework. Simulations and real-world tests demonstrate that the proposed PASC approach not only significantly boosts transmission efficiency, but also remains robust in diverse and evolving transmission scenarios.

eess.IV↗

arXiv2024

LLM-Slice: Dedicated Wireless Network Slicing for Large Language Models

The rapid adoption of large language models (LLMs) presents new challenges for existing network architectures due to significant peak traffic and high communication uncertainty. Traditional wireless networks struggle to support efficiently, leading to intolerable response delays, disconnections, and resource wastage. To address these issues, we propose LLM-Slice, the first system to provide dedicated communication slices for LLMs within a wireless network environment. By creating LLM-specific network slices, LLM-Slice efficiently binds services with communication resources. Based on user equipment (UE) requests and a permissions database, the system registers specific slices to offer controllable LLM services, integrating a downlink resource control module to optimize response speed, enhance resource utilization, and reduce disconnections. By deploying and validating in a real UE-gNB-CN environment, numerical results demonstrate that LLM-Slice significantly improves response speed and resource efficiency, providing a novel solution for fast and controllable LLM access in wireless networks.

cs.NI↗

arXiv2024

Why Does the Effective Context Length of LLMs Fall Short?

Advancements in distributed training and efficient attention mechanisms have significantly expanded the context window sizes of large language models (LLMs). However, recent work reveals that the effective context lengths of open-source LLMs often fall short, typically not exceeding half of their training lengths. In this work, we attribute this limitation to the left-skewed frequency distribution of relative positions formed in LLMs pretraining and post-training stages, which impedes their ability to effectively gather distant information. To address this challenge, we introduce ShifTed Rotray position embeddING (STRING). STRING shifts well-trained positions to overwrite the original ineffective positions during inference, enhancing performance within their existing training lengths. Experimental results show that without additional training, STRING dramatically improves the performance of the latest large-scale models, such as Llama3.1 70B and Qwen2 72B, by over 10 points on popular long-context benchmarks RULER and InfiniteBench, establishing new state-of-the-art results for open-source LLMs. Compared to commercial models, Llama 3.1 70B with \method even achieves better performance than GPT-4-128K and clearly surpasses Claude 2 and Kimi-chat.

cs.CL↗

arXiv2024

Higher-order and fractional discrete time crystals in Floquet-driven Rydberg atoms

Higher-order and fractional discrete time crystals (DTCs) are exotic phases of matter where the discrete time translation symmetry is broken into higher-order and non-integer category. Generation of these unique DTCs has been widely studied theoretically in different systems. However, no current experimental methods can probe these higher-order and fractional DTCs in any quantum many-body systems. We demonstrate an experimental approach to observe higher-order and fractional DTCs in Floquet-driven Rydberg atomic gases. We have discovered multiple $n$-DTCs with integer values of $n$ = 2, 3, and 4, and others ranging up to 14, along with fractional $n$-DTCs with $n$ values beyond the integers. The system response can transition between adjacent integer DTCs, during which the fractional DTCs are investigated. Study of higher-order and fractional DTCs expands fundamental knowledge of non-equilibrium dynamics and is promising for discovery of more complex temporal symmetries beyond the single discrete time translation symmetry.

cond-mat.quant-gas↗

arXiv2024

Graph Neural Network Enhanced Retrieval for Question Answering of LLMs

Retrieval augmented generation has revolutionized large language model (LLM) outputs by providing factual supports. Nevertheless, it struggles to capture all the necessary knowledge for complex reasoning questions. Existing retrieval methods typically divide reference documents into passages, treating them in isolation. These passages, however, are often interrelated, such as passages that are contiguous or share the same keywords. Therefore, it is crucial to recognize such relatedness for enhancing the retrieval process. In this paper, we propose a novel retrieval method, called GNN-Ret, which leverages graph neural networks (GNNs) to enhance retrieval by exploiting the relatedness between passages. Specifically, we first construct a graph of passages by connecting passages that are structure-related or keyword-related. A graph neural network (GNN) is then leveraged to exploit the relationships between passages and improve the retrieval of supporting passages. Furthermore, we extend our method to handle multi-hop reasoning questions using a recurrent graph neural network (RGNN), named RGNN-Ret. At each step, RGNN-Ret integrates the graphs of passages from previous steps, thereby enhancing the retrieval of supporting passages. Extensive experiments on benchmark datasets demonstrate that GNN-Ret achieves higher accuracy for question answering with a single query of LLMs than strong baselines that require multiple queries, and RGNN-Ret further improves accuracy and achieves state-of-the-art performance, with up to 10.4% accuracy improvement on the 2WikiMQA dataset.

cs.CL↗

arXiv2024

EmbodiedCity: A Benchmark Platform for Embodied Agent in Real-world City Environment

Embodied artificial intelligence emphasizes the role of an agent's body in generating human-like behaviors. The recent efforts on EmbodiedAI pay a lot of attention to building up machine learning models to possess perceiving, planning, and acting abilities, thereby enabling real-time interaction with the world. However, most works focus on bounded indoor environments, such as navigation in a room or manipulating a device, with limited exploration of embodying the agents in open-world scenarios. That is, embodied intelligence in the open and outdoor environment is less explored, for which one potential reason is the lack of high-quality simulators, benchmarks, and datasets. To address it, in this paper, we construct a benchmark platform for embodied intelligence evaluation in real-world city environments. Specifically, we first construct a highly realistic 3D simulation environment based on the real buildings, roads, and other elements in a real city. In this environment, we combine historically collected data and simulation algorithms to conduct simulations of pedestrian and vehicle flows with high fidelity. Further, we designed a set of evaluation tasks covering different EmbodiedAI abilities. Moreover, we provide a complete set of input and output interfaces for access, enabling embodied agents to easily take task requirements and current environmental observations as input and then make decisions and obtain performance evaluations. On the one hand, it expands the capability of existing embodied intelligence to higher levels. On the other hand, it has a higher practical value in the real world and can support more potential applications for artificial general intelligence. Based on this platform, we evaluate some popular large language models for embodied intelligence capabilities of different dimensions and difficulties.

cs.AI↗

arXiv2024

Kaleidoscope: Learnable Masks for Heterogeneous Multi-agent Reinforcement Learning

In multi-agent reinforcement learning (MARL), parameter sharing is commonly employed to enhance sample efficiency. However, the popular approach of full parameter sharing often leads to homogeneous policies among agents, potentially limiting the performance benefits that could be derived from policy diversity. To address this critical limitation, we introduce \emph{Kaleidoscope}, a novel adaptive partial parameter sharing scheme that fosters policy heterogeneity while still maintaining high sample efficiency. Specifically, Kaleidoscope maintains one set of common parameters alongside multiple sets of distinct, learnable masks for different agents, dictating the sharing of parameters. It promotes diversity among policy networks by encouraging discrepancy among these masks, without sacrificing the efficiencies of parameter sharing. This design allows Kaleidoscope to dynamically balance high sample efficiency with a broad policy representational capacity, effectively bridging the gap between full parameter sharing and non-parameter sharing across various environments. We further extend Kaleidoscope to critic ensembles in the context of actor-critic algorithms, which could help improve value estimations.Our empirical evaluations across extensive environments, including multi-agent particle environment, multi-agent MuJoCo and StarCraft multi-agent challenge v2, demonstrate the superior performance of Kaleidoscope compared with existing parameter sharing approaches, showcasing its potential for performance enhancement in MARL. The code is publicly available at \url{https://github.com/LXXXXR/Kaleidoscope}.

cs.LG↗