arXiv subjects

Jun Zhang

Publications and source records attributed to Jun Zhang.

At least 505 records · Page 28Linked to original sources

arXiv2024

GaussianImage: 1000 FPS Image Representation and Compression by 2D Gaussian Splatting

Implicit neural representations (INRs) recently achieved great success in image representation and compression, offering high visual quality and fast rendering speeds with 10-1000 FPS, assuming sufficient GPU resources are available. However, this requirement often hinders their use on low-end devices with limited memory. In response, we propose a groundbreaking paradigm of image representation and compression by 2D Gaussian Splatting, named GaussianImage. We first introduce 2D Gaussian to represent the image, where each Gaussian has 8 parameters including position, covariance and color. Subsequently, we unveil a novel rendering algorithm based on accumulated summation. Remarkably, our method with a minimum of 3$\times$ lower GPU memory usage and 5$\times$ faster fitting time not only rivals INRs (e.g., WIRE, I-NGP) in representation performance, but also delivers a faster rendering speed of 1500-2000 FPS regardless of parameter size. Furthermore, we integrate existing vector quantization technique to build an image codec. Experimental results demonstrate that our codec attains rate-distortion performance comparable to compression-based INRs such as COIN and COIN++, while facilitating decoding speeds of approximately 2000 FPS. Additionally, preliminary proof of concept shows that our codec surpasses COIN and COIN++ in performance when using partial bits-back coding. Code is available at https://github.com/Xinjie-Q/GaussianImage.

eess.IV↗

arXiv2024

Boosting Consistency in Story Visualization with Rich-Contextual Conditional Diffusion Models

Recent research showcases the considerable potential of conditional diffusion models for generating consistent stories. However, current methods, which predominantly generate stories in an autoregressive and excessively caption-dependent manner, often underrate the contextual consistency and relevance of frames during sequential generation. To address this, we propose a novel Rich-contextual Conditional Diffusion Models (RCDMs), a two-stage approach designed to enhance story generation's semantic consistency and temporal consistency. Specifically, in the first stage, the frame-prior transformer diffusion model is presented to predict the frame semantic embedding of the unknown clip by aligning the semantic correlations between the captions and frames of the known clip. The second stage establishes a robust model with rich contextual conditions, including reference images of the known clip, the predicted frame semantic embedding of the unknown clip, and text embeddings of all captions. By jointly injecting these rich contextual conditions at the image and feature levels, RCDMs can generate semantic and temporal consistency stories. Moreover, RCDMs can generate consistent stories with a single forward inference compared to autoregressive models. Our qualitative and quantitative results demonstrate that our proposed RCDMs outperform in challenging scenarios. The code and model will be available at https://github.com/muzishen/RCDMs.

cs.CV↗

arXiv2024

Accurate Shear Recovery with Multi-Band Images of Hyper Suprime-Cam

The existing large scale weak lensing surveys typically reserve the best seeing conditions for a certain optical band to minimize shape measurement errors and maximize the number of usable background galaxies. This is because most popular shear measurement methods contain explicit or implicit thresholds on the galaxy-to-PSF (point spread function) size ratio, below which their shape measurement errors increase abruptly. Using the DECaLS data, we have previously demonstrated that the Fourier\_Quad method performs very well on poorly resolved galaxy images in general. It is therefore a ready tool for shear measurement with multi-band images regardless of their seeing conditions. In this paper, we apply the Fourier\_Quad pipeline on the multi-band images from the third public data release of the Hyper Suprime-Cam Subaru Strategic Program. We show that the shear catalogs from the five optical bands (g/r/i/z/y) all pass the field-distortion test with very high accuracy. Using the LOWZ and CMASS galaxies as foreground lenses, we show that the errorbar in the galaxy-galaxy lensing measurement can be decreased by factors around 15\% by combining shear catalogs from different bands. This indicates that it is worthful to do multi-bands shear measurements for a better shear statistics.

astro-ph.CO↗

arXiv2024

Tele-Correlation: Calibrating Shear-Shear Correlation with Real Data

Tele-correlation refers to the correlation of galaxy shapes with large angular separations (e.g., $>100$ degrees). Since there are no astrophysical reasons causing such a correlation on cosmological scales, any detected tele-correlation could disclose systematic effects in shear-shear correlation measurement. If the shear estimators are measured on single exposures, we show that the field distortion (FD) signal associated with the galaxy position on the CCD can be retained and used in tele-correlation to help us directly calibrate the multiplicative and additive biases in shear-shear correlations. We use the DECaLS shear catalog produced by the Fourier\_Quad pipeline to demonstrate this idea. To our surprise, we find that significant multiplicative biases can arise (up to more than 10\%) due to redshift binning of the galaxies. Correction for this bias leads to about 1$σ$ increase of the best-fit value of $S_8$ from $0.760^{+0.015}_{-0.017}$ to $0.777^{+0.016}_{-0.019}$ in our tomography study.

astro-ph.CO↗

arXiv2024

A Comprehensive Solution to Connect Speech Encoder and Large Language Model for ASR

Recent works have shown promising results in connecting speech encoders to large language models (LLMs) for speech recognition. However, several limitations persist, including limited fine-tuning options, a lack of mechanisms to enforce speech-text alignment, and high insertion errors especially in domain mismatch conditions. This paper presents a comprehensive solution to address these issues. We begin by investigating more thoughtful fine-tuning schemes. Next, we propose a matching loss to enhance alignment between modalities. Finally, we explore training and inference methods to mitigate high insertion errors. Experimental results on the Librispeech corpus demonstrate that partially fine-tuning the encoder and LLM using parameter-efficient methods, such as LoRA, is the most cost-effective approach. Additionally, the matching loss improves modality alignment, enhancing performance. The proposed training and inference methods significantly reduce insertion errors.

cs.LG↗

arXiv2024

Low-Complexity CSI Feedback for FDD Massive MIMO Systems via Learning to Optimize

In frequency-division duplex (FDD) massive multiple-input multiple-output (MIMO) systems, the growing number of base station antennas leads to prohibitive feedback overhead for downlink channel state information (CSI). To address this challenge, state-of-the-art (SOTA) fully data-driven deep learning (DL)-based CSI feedback schemes have been proposed. However, the high computational complexity and memory requirements of these methods hinder their practical deployment on resource-constrained devices like mobile phones. To solve the problem, we propose a model-driven DL-based CSI feedback approach by integrating the wisdom of compressive sensing and learning to optimize (L2O). Specifically, only a linear learnable projection is adopted at the encoder side to compress the CSI matrix, thereby significantly cutting down the user-side complexity and memory expenditure. On the other hand, the decoder incorporates two specially designed components, i.e., a learnable sparse transformation and an element-wise L2O reconstruction module. The former is developed to learn a sparse basis for CSI within the angular domain, which explores channel sparsity effectively. The latter shares the same long short term memory (LSTM) network across all elements of the optimization variable, eliminating the retraining cost when problem scale changes. Simulation results show that the proposed method achieves a comparable performance with the SOTA CSI feedback scheme but with much-reduced complexity, and enables multiple-rate feedback.

eess.SP↗

arXiv2024

A Bayesian Drift-Diffusion Model of Schachter-Singer's Two Factor Theory of Emotion

Bayesian inference has been used in the past to model visual perception (Kersten, Mamassian, & Yuille, 2004), accounting for the Helmholtz principle of perception as "unconscious inference" that is constrained by bottom-up sensory evidence (likelihood) while subject to top-down expectation, priming, or other contextual influences (prior bias); here "unconsciousness" merely relates to the "directness" of perception in the sense of Gibson. Here, we adopt the same Bayesian framework to model emotion process in accordance with Schachter-Singer's Two-Factor theory, which argues that emotion is the outcome of cognitive labeling or attribution of a diffuse pattern of autonomic arousal (Schachter & Singer, 1962). In analogous to visual perception, we conceptualize the emotion process as an instance of Bayesian inference, combining the contextual information with a person's physiological arousal patterns. Drift-diffusion models were constructed to simulate emotional processes, where the decision boundaries correspond to the emotional state experienced by the participants, and boundary-crossing constitutes "labeling" in Schachter-Singer's sense. Our model is tested against experimental data from the Schachter & Singer's study (1962) and the Ross et al. study (1969). Two model scenarios are investigated, in which arousal pattern as one factor is pitted against contextual interaction with an confederate (in Schachter-Singer case) or explicitly instructed mis-attribution (in Ross et al. case) as another factor, mapping onto the Bayesian prior (initial position of the drift) and the likelihood function (evidence accumulation or drift rate). We find that the first scenario (arousal as the prior and context as the likelihood) has a better fit with Schachter & Singer (1962) whereas the second scenario (context as the prior and arousal as the likelihood) has a better fit with Ross et al. (1969).

cs.CE↗

arXiv2024

Bayes-Optimal Unsupervised Learning for Channel Estimation in Near-Field Holographic MIMO

Holographic MIMO (HMIMO) is being increasingly recognized as a key enabling technology for 6G wireless systems through the deployment of an extremely large number of antennas within a compact space to fully exploit the potentials of the electromagnetic (EM) channel. Nevertheless, the benefits of HMIMO systems cannot be fully unleashed without an efficient means to estimate the high-dimensional channel, whose distribution becomes increasingly complicated due to the accessibility of the near-field region. In this paper, we address the fundamental challenge of designing a low-complexity Bayes-optimal channel estimator in near-field HMIMO systems operating in unknown EM environments. The core idea is to estimate the HMIMO channels solely based on the Stein's score function of the received pilot signals and an estimated noise level, without relying on priors or supervision that is not feasible in practical deployment. A neural network is trained with the unsupervised denoising score matching objective to learn the parameterized score function. Meanwhile, a principal component analysis (PCA)-based algorithm is proposed to estimate the noise level leveraging the low-rank near-field spatial correlation. Building upon these techniques, we develop a Bayes-optimal score-based channel estimator for fully-digital HMIMO transceivers in a closed form. The optimal score-based estimator is also extended to hybrid analog-digital HMIMO systems by incorporating it into a low-complexity message passing algorithm. The (quasi-) Bayes-optimality of the proposed estimators is validated both in theory and by extensive simulation results. In addition to optimality, it is shown that our proposal is robust to various mismatches and can quickly adapt to dynamic EM environments in an online manner thanks to its unsupervised nature, demonstrating its potential in real-world deployment.

eess.SP↗

arXiv2024

WirelessLLM: Empowering Large Language Models Towards Wireless Intelligence

The rapid evolution of wireless technologies and the growing complexity of network infrastructures necessitate a paradigm shift in how communication networks are designed, configured, and managed. Recent advancements in Large Language Models (LLMs) have sparked interest in their potential to revolutionize wireless communication systems. However, existing studies on LLMs for wireless systems are limited to a direct application for telecom language understanding. To empower LLMs with knowledge and expertise in the wireless domain, this paper proposes WirelessLLM, a comprehensive framework for adapting and enhancing LLMs to address the unique challenges and requirements of wireless communication networks. We first identify three foundational principles that underpin WirelessLLM: knowledge alignment, knowledge fusion, and knowledge evolution. Then, we investigate the enabling technologies to build WirelessLLM, including prompt engineering, retrieval augmented generation, tool usage, multi-modal pre-training, and domain-specific fine-tuning. Moreover, we present three case studies to demonstrate the practical applicability and benefits of WirelessLLM for solving typical problems in wireless networks. Finally, we conclude this paper by highlighting key challenges and outlining potential avenues for future research.

cs.NI↗

arXiv2024

Can Large Language Models Understand Spatial Audio?

This paper explores enabling large language models (LLMs) to understand spatial information from multichannel audio, a skill currently lacking in auditory LLMs. By leveraging LLMs' advanced cognitive and inferential abilities, the aim is to enhance understanding of 3D environments via audio. We study 3 spatial audio tasks: sound source localization (SSL), far-field speech recognition (FSR), and localisation-informed speech extraction (LSE), achieving notable progress in each task. For SSL, our approach achieves an MAE of $2.70^{\circ}$ on the Spatial LibriSpeech dataset, substantially surpassing the prior benchmark of about $6.60^{\circ}$. Moreover, our model can employ spatial cues to improve FSR accuracy and execute LSE by selectively attending to sounds originating from a specified direction via text prompts, even amidst overlapping speech. These findings highlight the potential of adapting LLMs to grasp physical audio concepts, paving the way for LLM-based agents in 3D environments.

cs.SD↗

arXiv2024

Living in the Moment: Can Large Language Models Grasp Co-Temporal Reasoning?

Temporal reasoning is fundamental for large language models (LLMs) to comprehend the world. Current temporal reasoning datasets are limited to questions about single or isolated events, falling short in mirroring the realistic temporal characteristics involving concurrent nature and intricate temporal interconnections. In this paper, we introduce CoTempQA, a comprehensive co-temporal Question Answering (QA) benchmark containing four co-temporal scenarios (Equal, Overlap, During, Mix) with 4,748 samples for evaluating the co-temporal comprehension and reasoning abilities of LLMs. Our extensive experiments reveal a significant gap between the performance of current LLMs and human-level reasoning on CoTempQA tasks. Even when enhanced with Chain of Thought (CoT) methodologies, models consistently struggle with our task. In our preliminary exploration, we discovered that mathematical reasoning plays a significant role in handling co-temporal events and proposed a strategy to boost LLMs' co-temporal reasoning from a mathematical perspective. We hope that our CoTempQA datasets will encourage further advancements in improving the co-temporal reasoning capabilities of LLMs. Our code is available at https://github.com/zhaochen0110/Cotempqa.

cs.CL↗

arXiv2024

Dual-Pipeline with Low-Rank Adaptation for New Language Integration in Multilingual ASR

This paper addresses challenges in integrating new languages into a pre-trained multilingual automatic speech recognition (mASR) system, particularly in scenarios where training data for existing languages is limited or unavailable. The proposed method employs a dual-pipeline with low-rank adaptation (LoRA). It maintains two data flow pipelines-one for existing languages and another for new languages. The primary pipeline follows the standard flow through the pre-trained parameters of mASR, while the secondary pipeline additionally utilizes language-specific parameters represented by LoRA and a separate output decoder module. Importantly, the proposed approach minimizes the performance degradation of existing languages and enables a language-agnostic operation mode, facilitated by a decoder selection strategy. We validate the effectiveness of the proposed method by extending the pre-trained Whisper model to 19 new languages from the FLEURS dataset

eess.AS↗

arXiv2024

A Federated Online Restless Bandit Framework for Cooperative Resource Allocation

Restless multi-armed bandits (RMABs) have been widely utilized to address resource allocation problems with Markov reward processes (MRPs). Existing works often assume that the dynamics of MRPs are known prior, which makes the RMAB problem solvable from an optimization perspective. Nevertheless, an efficient learning-based solution for RMABs with unknown system dynamics remains an open problem. In this paper, we study the cooperative resource allocation problem with unknown system dynamics of MRPs. This problem can be modeled as a multi-agent online RMAB problem, where multiple agents collaboratively learn the system dynamics while maximizing their accumulated rewards. We devise a federated online RMAB framework to mitigate the communication overhead and data privacy issue by adopting the federated learning paradigm. Based on this framework, we put forth a Federated Thompson Sampling-enabled Whittle Index (FedTSWI) algorithm to solve this multi-agent online RMAB problem. The FedTSWI algorithm enjoys a high communication and computation efficiency, and a privacy guarantee. Moreover, we derive a regret upper bound for the FedTSWI algorithm. Finally, we demonstrate the effectiveness of the proposed algorithm on the case of online multi-user multi-channel access. Numerical results show that the proposed algorithm achieves a fast convergence rate of $\mathcal{O}(\sqrt{T\log(T)})$ and better performance compared with baselines. More importantly, its sample complexity decreases with the number of agents.

cs.LG↗

arXiv2024

On primordial universe in anti-de Sitter landscape

How the spacetime evolved non-perturbatively in a landscape with multiple anti-de Sitter (AdS) vacua, which is theoretically well-motivated, has always been a matter of concern. As a step towards this issue, we perform (3+1)D numerical relativity simulations for the inhomogeneous universe in an AdS landscape, and find that large inhomogeneity of scalar field can develop into not only sphere-like bubbles, but also novel tube-like structures. It is observed that the bubble or tube wall (across the potential barrier) likely inflates as a quasi-dS space, while the different regions separated by the walls are in different AdS vacua and will collapse towards singularity.

gr-qc↗

arXiv2024

Memorization in deep learning: A survey

Deep Learning (DL) powered by Deep Neural Networks (DNNs) has revolutionized various domains, yet understanding the intricacies of DNN decision-making and learning processes remains a significant challenge. Recent investigations have uncovered an interesting memorization phenomenon in which DNNs tend to memorize specific details from examples rather than learning general patterns, affecting model generalization, security, and privacy. This raises critical questions about the nature of generalization in DNNs and their susceptibility to security breaches. In this survey, we present a systematic framework to organize memorization definitions based on the generalization and security/privacy domains and summarize memorization evaluation methods at both the example and model levels. Through a comprehensive literature review, we explore DNN memorization behaviors and their impacts on security and privacy. We also introduce privacy vulnerabilities caused by memorization and the phenomenon of forgetting and explore its connection with memorization. Furthermore, we spotlight various applications leveraging memorization and forgetting mechanisms, including noisy label learning, privacy preservation, and model enhancement. This survey offers the first-in-kind understanding of memorization in DNNs, providing insights into its challenges and opportunities for enhancing AI development while addressing critical ethical concerns.

cs.LG↗

arXiv2024

DA-HFNet: Progressive Fine-Grained Forgery Image Detection and Localization Based on Dual Attention

The increasing difficulty in accurately detecting forged images generated by AIGC(Artificial Intelligence Generative Content) poses many risks, necessitating the development of effective methods to identify and further locate forged areas. In this paper, to facilitate research efforts, we construct a DA-HFNet forged image dataset guided by text or image-assisted GAN and Diffusion model. Our goal is to utilize a hierarchical progressive network to capture forged artifacts at different scales for detection and localization. Specifically, it relies on a dual-attention mechanism to adaptively fuse multi-modal image features in depth, followed by a multi-branch interaction network to thoroughly interact image features at different scales and improve detector performance by leveraging dependencies between layers. Additionally, we extract more sensitive noise fingerprints to obtain more prominent forged artifact features in the forged areas. Extensive experiments validate the effectiveness of our approach, demonstrating significant performance improvements compared to state-of-the-art methods for forged image detection and localization.The code and dataset will be released in the future.

cs.CV↗

arXiv2024

V-Express: Conditional Dropout for Progressive Training of Portrait Video Generation

In the field of portrait video generation, the use of single images to generate portrait videos has become increasingly prevalent. A common approach involves leveraging generative models to enhance adapters for controlled generation. However, control signals (e.g., text, audio, reference image, pose, depth map, etc.) can vary in strength. Among these, weaker conditions often struggle to be effective due to interference from stronger conditions, posing a challenge in balancing these conditions. In our work on portrait video generation, we identified audio signals as particularly weak, often overshadowed by stronger signals such as facial pose and reference image. However, direct training with weak signals often leads to difficulties in convergence. To address this, we propose V-Express, a simple method that balances different control signals through the progressive training and the conditional dropout operation. Our method gradually enables effective control by weak conditions, thereby achieving generation capabilities that simultaneously take into account the facial pose, reference image, and audio. The experimental results demonstrate that our method can effectively generate portrait videos controlled by audio. Furthermore, a potential solution is provided for the simultaneous and effective use of conditions of varying strengths.

cs.CV↗

arXiv2024

Freeplane: Unlocking Free Lunch in Triplane-Based Sparse-View Reconstruction Models

Creating 3D assets from single-view images is a complex task that demands a deep understanding of the world. Recently, feed-forward 3D generative models have made significant progress by training large reconstruction models on extensive 3D datasets, with triplanes being the preferred 3D geometry representation. However, effectively utilizing the geometric priors of triplanes, while minimizing artifacts caused by generated inconsistent multi-view images, remains a challenge. In this work, we present \textbf{Fre}quency modulat\textbf{e}d tri\textbf{plane} (\textbf{Freeplane}), a simple yet effective method to improve the generation quality of feed-forward models without additional training. We first analyze the role of triplanes in feed-forward methods and find that the inconsistent multi-view images introduce high-frequency artifacts on triplanes, leading to low-quality 3D meshes. Based on this observation, we propose strategically filtering triplane features and combining triplanes before and after filtering to produce high-quality textured meshes. These techniques incur no additional cost and can be seamlessly integrated into pre-trained feed-forward models to enhance their robustness against the inconsistency of generated multi-view images. Both qualitative and quantitative results demonstrate that our method improves the performance of feed-forward models by simply modulating triplanes. All you need is to modulate the triplanes during inference.

cs.CV↗