Search

Retrieval-augmented Large Language Models (LLMs) have reshaped traditional query-answering systems, offering unparalleled user experiences. However, existing retrieval techniques often struggle to handle multi-modal query contexts. In this paper, we present an interactive Multi-modal Query Answering (MQA) system, empowered by our newly developed multi-modal retrieval framework and navigation graph index, integrated with cutting-edge LLMs. It comprises five core components: Data Preprocessing, Vector Representation, Index Construction, Query Execution, and Answer Generation, all orchestrated by a dedicated coordinator to ensure smooth data flow from input to answer generation. One notable aspect of MQA is its utilization of contrastive learning to assess the significance of different modalities, facilitating precise measurement of multi-modal information similarity. Furthermore, the system achieves efficient retrieval through our advanced navigation graph index, refined using computational pruning techniques. Another highlight of our system is its pluggable processing framework, allowing seamless integration of embedding models, graph indexes, and LLMs. This flexibility provides users diverse options for gaining insights from their multi-modal knowledge base. A preliminary video introduction of MQA is available at https://youtu.be/xvUuo2ZIqWk., Comment: This demo paper has been accepted by VLDB 2024

Rooting in the scarcity of most attributes, realistic pedestrian attribute datasets exhibit unduly skewed data distribution, from which two types of model failures are delivered: (1) label imbalance: model predictions lean greatly towards the side of majority labels; (2) semantics imbalance: model is easily overfitted on the under-represented attributes due to their insufficient semantic diversity. To render perfect label balancing, we propose a novel framework that successfully decouples label-balanced data re-sampling from the curse of attributes co-occurrence, i.e., we equalize the sampling prior of an attribute while not biasing that of the co-occurred others. To diversify the attributes semantics and mitigate the feature noise, we propose a Bayesian feature augmentation method to introduce true in-distribution novelty. Handling both imbalances jointly, our work achieves best accuracy on various popular benchmarks, and importantly, with minimal computational budget., Comment: Accepted as ICML2024 main conference paper

Acquiring and utilizing accurate channel state information (CSI) can significantly improve transmission performance, thereby holding a crucial role in realizing the potential advantages of massive multiple-input multiple-output (MIMO) technology. Current prevailing CSI feedback approaches improve precision by employing advanced deep-learning methods to learn representative CSI features for a subsequent compression process. Diverging from previous works, we treat the CSI compression problem in the context of implicit neural representations. Specifically, each CSI matrix is viewed as a neural function that maps the CSI coordinates (antenna number and subchannel) to the corresponding channel gains. Instead of transmitting the parameters of the implicit neural functions directly, we transmit modulations based on the CSI matrix derived through a meta-learning algorithm. Modulations are then applied to a shared base network to generate the elements of the CSI matrix. Modulations corresponding to the CSI matrix are quantized and entropy-coded to further reduce the communication bandwidth, thus achieving extreme CSI compression ratios. Numerical results show that our proposed approach achieves state-of-the-art performance and showcases flexibility in feedback strategies.

This paper introduces an innovative deep joint source-channel coding (DeepJSCC) approach to image transmission over a cooperative relay channel. The relay either amplifies and forwards a scaled version of its received signal, referred to as DeepJSCC-AF, or leverages neural networks to extract relevant features about the source signal before forwarding it to the destination, which we call DeepJSCC-PF (Process-and-Forward). In the full-duplex scheme, inspired by the block Markov coding (BMC) concept, we introduce a novel block transmission strategy built upon novel vision transformer architecture. In the proposed scheme, the source transmits information in blocks, and the relay updates its knowledge about the input signal after each block and generates its own signal to be conveyed to the destination. To enhance practicality, we introduce an adaptive transmission model, which allows a single trained DeepJSCC model to adapt seamlessly to various channel qualities, making it a versatile solution. Simulation results demonstrate the superior performance of our proposed DeepJSCC compared to the state-of-the-art BPG image compression algorithm, even when operating at the maximum achievable rate of conventional decode-and-forward and compress-and-forward protocols, for both half-duplex and full-duplex relay scenarios., Comment: Submitted for possible IEEE journal

Kr\"oncke has shown that the Fubini-Study metric is an unstable generalized stationary solution of Ricci flow [Kr\"o20]. In this paper, we carry out numerical simulations which indicate that Ricci flow solutions originating at unstable perturbations of the Fubini-Study metric develop local singularities modeled by the blowdown soliton discovered in [FIK03]., Comment: 14 pages, 3 figures. Comments are welcome!

knowledge graph-based recommendation methods have achieved great success in the field of recommender systems. However, over-reliance on high-quality knowledge graphs is a bottleneck for such methods. Specifically, the long-tailed distribution of entities of KG and noise issues in the real world will make item-entity dependent relations deviate from reflecting true characteristics and significantly harm the performance of modeling user preference. Contrastive learning, as a novel method that is employed for data augmentation and denoising, provides inspiration to fill this research gap. However, the mainstream work only focuses on the long-tail properties of the number of items clicked, while ignoring that the long-tail properties of total number of clicks per user may also affect the performance of the recommendation model. Therefore, to tackle these problems, motivated by the Debiased Contrastive Learning of Unsupervised Sentence Representations (DCLR), we propose Two-Level Debiased Contrastive Graph Learning (TDCGL) model. Specifically, we design the Two-Level Debiased Contrastive Learning (TDCL) and deploy it in the KG, which is conducted not only on User-Item pairs but also on User-User pairs for modeling higher-order relations. Also, to reduce the bias caused by random sampling in contrastive learning, with the exception of the negative samples obtained by random sampling, we add a noise-based generation of negation to ensure spatial uniformity. Considerable experiments on open-source datasets demonstrate that our method has excellent anti-noise capability and significantly outperforms state-of-the-art baselines. In addition, ablation studies about the necessity for each level of TDCL are conducted.

This paper introduces a vision transformer (ViT)-based deep joint source and channel coding (DeepJSCC) scheme for wireless image transmission over multiple-input multiple-output (MIMO) channels, denoted as DeepJSCC-MIMO. We consider DeepJSCC-MIMO for adaptive image transmission in both open-loop and closed-loop MIMO systems. The novel DeepJSCC-MIMO architecture surpasses the classical separation-based benchmarks with robustness to channel estimation errors and showcases remarkable flexibility in adapting to diverse channel conditions and antenna numbers without requiring retraining. Specifically, by harnessing the self-attention mechanism of ViT, DeepJSCC-MIMO intelligently learns feature mapping and power allocation strategies tailored to the unique characteristics of the source image and prevailing channel conditions. Extensive numerical experiments validate the significant improvements in transmission quality achieved by DeepJSCC-MIMO for both open-loop and closed-loop MIMO systems across a wide range of scenarios. Moreover, DeepJSCC-MIMO exhibits robustness to varying channel conditions, channel estimation errors, and different antenna numbers, making it an appealing solution for emerging semantic communication systems., Comment: arXiv admin note: text overlap with arXiv:2210.15347

This paper presents a novel wireless image transmission paradigm that can exploit feedback from the receiver, called DeepJSCC-ViT-f. We consider a block feedback channel model, where the transmitter receives noiseless/noisy channel output feedback after each block. The proposed scheme employs a single encoder to facilitate transmission over multiple blocks, refining the receiver's estimation at each block. Specifically, the unified encoder of DeepJSCC-ViT-f can leverage the semantic information from the source image, and acquire channel state information and the decoder's current belief about the source image from the feedback signal to generate coded symbols at each block. Numerical experiments show that our DeepJSCC-ViT-f scheme achieves state-of-the-art transmission performance with robustness to noise in the feedback link. Additionally, DeepJSCC-ViT-f can adapt to the channel condition directly through feedback without the need for separate channel estimation. We further extend the scope of the DeepJSCC-ViT-f approach to include the broadcast channel, which enables the transmitter to generate broadcast codes in accordance with signal semantics and channel feedback from individual receivers.

To improve the performance in identifying the faults under strong noise for rotating machinery, this paper presents a dynamic feature reconstruction signal graph method, which plays the key role of the proposed end-to-end fault diagnosis model. Specifically, the original mechanical signal is first decomposed by wavelet packet decomposition (WPD) to obtain multiple subbands including coefficient matrix. Then, with originally defined two feature extraction factors MDD and DDD, a dynamic feature selection method based on L2 energy norm (DFSL) is proposed, which can dynamically select the feature coefficient matrix of WPD based on the difference in the distribution of norm energy, enabling each sub-signal to take adaptive signal reconstruction. Next the coefficient matrices of the optimal feature sub-bands are reconstructed and reorganized to obtain the feature signal graphs. Finally, deep features are extracted from the feature signal graphs by 2D-Convolutional neural network (2D-CNN). Experimental results on a public data platform of a bearing and our laboratory platform of robot grinding show that this method is better than the existing methods under different noise intensities., Comment: The feature extraction algorithm DFSL has errors in derivation and experimental deficiencies

We study the collaborative image retrieval problem at the wireless edge, where multiple edge devices capture images of the same object, which are then used jointly to retrieve similar images at the edge server over a shared multiple access channel. We propose a semantic non-orthogonal multiple access (NOMA) communication paradigm, in which extracted features from each device are mapped directly to channel inputs, which are then added over-the-air. We propose a novel contrastive learning (CL)-based semantic communication (CL-SC) paradigm, aiming to exploit signal correlations to maximize the retrieval accuracy under a total bandwidth constraints. Specifically, we treat noisy correlated signals as different augmentations of a common identity, and propose a cross-view CL algorithm to optimize the correlated signals in a coarse-to-fine fashion to improve retrieval accuracy. Extensive numerical experiments verify that our method achieves the state-of-the-art performance and can significantly improve retrieval accuracy, with particularly significant gains in low signla-to-noise ratio (SNR) and limited bandwidth regimes.

We study the collaborative image retrieval problem at the wireless edge, where multiple edge devices capture images of the same object from different angles and locations, which are then used jointly to retrieve similar images at the edge server over a shared multiple access channel (MAC). We propose two novel deep learning-based joint source and channel coding (JSCC) schemes for the task over both additive white Gaussian noise (AWGN) and Rayleigh slow fading channels, with the aim of maximizing the retrieval accuracy under a total bandwidth constraint. The proposed schemes are evaluated on a wide range of channel signal-to-noise ratios (SNRs), and shown to outperform the single-device JSCC and the separation-based multiple-access benchmarks. We also propose two novel SNR-aware JSCC schemes with attention modules to improve the performance in the case of channel mismatch between training and test instances.

This paper presents a novel deep joint source-channel coding (DeepJSCC) scheme for image transmission over a half-duplex cooperative relay channel. Specifically, we apply DeepJSCC to two basic modes of cooperative communications, namely amplify-and-forward (AF) and decode-and-forward (DF). In DeepJSCC-AF, the relay simply amplifies and forwards its received signal. In DeepJSCC-DF, on the other hand, the relay first reconstructs the transmitted image and then re-encodes it before forwarding. Considering the excessive computation overhead of DeepJSCC-DF for recovering the image at the relay, we propose an alternative scheme, called DeepJSCC-PF, in which the relay processes and forwards its received signal without necessarily recovering the image. Simulation results show that the proposed DeepJSCC-AF, DF, and PF schemes are superior to the digital baselines with BPG compression with polar codes and provides a graceful performance degradation with deteriorating channel quality. Further investigation shows that the PSNR gain of DeepJSCC-DF/PF over DeepJSCC-AF improves as the channel condition between the source and relay improves. Moreover, DeepJSCC-PF scheme achieves a similar performance to DeepJSCC-DF with lower computational complexity., Comment: Accepted to IEEE International Conference on Machine Learning for Communication and Networking (ICMLCN) 2024, code available via this link https://github.com/aprilbian/Relay_JSCC

ASR can be improved by multi-task learning (MTL) with domain enhancing or domain adversarial training, which are two opposite objectives with the aim to increase/decrease domain variance towards domain-aware/agnostic ASR, respectively. In this work, we study how to best apply these two opposite objectives with speaker labels to improve conformer-based ASR. We also propose a novel adaptive gradient reversal layer for stable and effective adversarial training without tuning effort. Detailed analysis and experimental verification are conducted to show the optimal positions in the ASR neural network (NN) to apply speaker enhancing and adversarial training. We also explore their combination for further improvement, achieving the same performance as i-vectors plus adversarial training. Our best speaker-based MTL achieves 7\% relative improvement on the Switchboard Hub5'00 set. We also investigate the effect of such speaker-based MTL w.r.t. cleaner dataset and weaker ASR NN., Comment: accepted at ICASSP 2023

We propose novel deep joint source-channel coding (DeepJSCC) algorithms for wireless image transmission over multi-input multi-output (MIMO) Rayleigh fading channels, when channel state information (CSI) is available only at the receiver. We consider two different schemes; one exploiting the spatial diversity and the other exploiting the spatial multiplexing gain of the MIMO channel, respectively. For the former, we utilize an orthogonal space-time block code (OSTBC) to achieve full diversity and increase the robustness against channel variations. In the latter, we directly map the input to the antennas, where the additional degree of freedom can be used to send more information about the source signal. Simulation results show that the diversity scheme outperforms the multiplexing scheme for lower signal-to-noise ratio (SNR) values and a smaller number of receive antennas at the AP. When the number of transmit antennas is greater than two, however, the full-diversity scheme becomes less beneficial. We also show that both the diversity and multiplexing schemes can achieve comparable performance with the state-of-the-art BPG algorithm delivered at the instantaneous capacity of the MIMO channel, which serves as an upper bound on the performance of separation-based practical systems., Comment: Accepted to SPAWC 2023, 5 pages

This paper presents a vision transformer (ViT) based joint source and channel coding (JSCC) scheme for wireless image transmission over multiple-input multiple-output (MIMO) systems, called ViT-MIMO. The proposed ViT-MIMO architecture, in addition to outperforming separation-based benchmarks, can flexibly adapt to different channel conditions without requiring retraining. Specifically, exploiting the self-attention mechanism of the ViT enables the proposed ViT-MIMO model to adaptively learn the feature mapping and power allocation based on the source image and channel conditions. Numerical experiments show that ViT-MIMO can significantly improve the transmission quality cross a large variety of scenarios, including varying channel conditions, making it an attractive solution for emerging semantic communication systems.

We improve the understanding of both finite time and infinite time singularities of the modified K\"ahler-Ricci flow as initiated by the second author of this paper in [26]. This is done by relating the modified K\"ahler-Ricci flow with the recent studies on the classic K\"ahler-Ricci flow and the degenerate complex Monge-Amp\`ere equation., Comment: 20 pages. Comments welcome

Detecting manipulated facial images and videos on social networks has been an urgent problem to be solved. The compression of videos on social media has destroyed some pixel details that could be used to detect forgeries. Hence, it is crucial to detect manipulated faces in videos of different quality. We propose a new multi-stream network architecture named GGViT, which utilizes global information to improve the generalization of the model. The embedding of the whole face extracted by ViT will guide each stream network. Through a large number of experiments, we have proved that our proposed model achieves state-of-the-art classification accuracy on FF++ dataset, and has been greatly improved on scenarios of different compression rates. The accuracy of Raw/C23, Raw/C40 and C23/C40 was increased by 24.34%, 15.08% and 10.14% respectively., Comment: 6 pages,4 figures,to be published in ICPR2022

In this paper, we prove a quantitative spectral inequality for the second Robin eigenvalue in non-compact rank-1 symmetric spaces. In particular, this shows that for bounded domains in non-compact rank-1 symmetric spaces, the geodesic ball maximises the second Robin eigenvalue among domains of the same volume, with negative Robin parameter in the regime connecting the first nontrivial Neumann and Steklov eigenvalues. This result generalises the work of Freitas and Laugesen in the Euclidean setting [FL21] as well as our previous work in the hyperbolic space [LWW20]., Comment: 16 pages. Comments are welcome. arXiv admin note: substantial text overlap with arXiv:2003.03087

In previous work [GIKW21], we have presented evidence from numerical simulations that the Type-II singularities of mean curvature flow (MCF) of rotationally-symmetric, complete, noncompact embedded hypersurfaces constructed in [IW19, IWZ21] are stable. More precisely, it is shown in that paper that for small rotationally-symmetric perturbations of initial embeddings near the "tip", numerical simulations of MCF of such initial embeddings develop the same Type-II singularities with the same "bowl soliton" blowup behaviors in a neighborhood of the singularity. It is also shown in that work that for small rotationally-symmetric perturbations of the initial embeddings that are sufficiently far away from the tip, MCF develops Type-I "neckpinch" singularities. In this work, we again use numerical simulations to show that MCF subject to initial perturbations that are not rotationally symmetric behaves asymptotically like it does for rotationally-symmetric perturbations. In particular, if we impose sinusoidal angular dependence on the initial embeddings, we find that for perturbations near the tip, evolutions by MCF asymptotically lose their angular dependence -- becoming round -- and develop Type-II bowl soliton singularities. As well, if we impose sinusoidal angular dependence on the initial embeddings for perturbations sufficiently far from the tip, the angular dependence again disappears as Type-I neckpinch singularities develop. The numerical analysis carried out in this work is an adaptation of the "overlap" method introduced in [GIKW21] and permits angular dependence., Comment: 17 pages, 13 figures. Comments are welcome!

Customer churn prediction is a challenging domain of research that contributes to customer retention strategy. The predictive performance of existing machine learning models, which are often adopted by churn communities, appear to be at a bottleneck, partly due to models' poor feature extraction capability. Therefore, a novel algorithm, a hybrid neural network with self-attention enhancement (HNNSAE), is proposed in this paper to improve the efficiency of feature screening and feature extraction, consequently improving the model's predictive performance. This model consists of three main blocks. The first block is the entity embedding layer, which is employed to process the categorical variables transformed into 0-1 code. The second block is the feature extractor, which extracts the significant features through the multi-head self-attention mechanism. In addition, to improve the feature extraction effect, we stack the residual connection neural network on multi-head self-attention modules. The third block is a classifier, which is a three-layer multilayer perceptron. This work conducts experiments on publicly available dataset related to commercial bank customers. The result demonstrates that HNNSAE significantly outperforms the other Individual Machine Learning (IML), Ensemble Machine Learning (EML), and Deep Learning (DL) methods tested in this paper. Furthermore, we compare the performance of the feature extractor proposed in this paper with that of other three feature extractors and find that the method proposed in this paper significantly outperforms other methods. In addition, four hypotheses about model prediction performance and overfitting risk are tested on the publicly available dataset., Comment: 34 pages, 12 figures

We present a learning-based channel-adaptive joint source and channel coding (CA-JSCC) scheme for wireless image transmission over multipath fading channels. The proposed method is an end-to-end autoencoder architecture with a dual-attention mechanism employing orthogonal frequency division multiplexing (OFDM) transmission. Unlike the previous works, our approach is adaptive to channel-gain and noise-power variations by exploiting the estimated channel state information (CSI). Specifically, with the proposed dual-attention mechanism, our model can learn to map the features and allocate transmission-power resources judiciously based on the estimated CSI. Extensive numerical experiments verify that CA-JSCC achieves state-of-the-art performance among existing JSCC schemes. In addition, CA-JSCC is robust to varying channel conditions and can better exploit the limited channel resources by transmitting critical features over better subchannels., Comment: IEEE Wireless Communications Letters

This paper presents a self-supervised feature learning method for hyperspectral image classification. Our method tries to construct two different views of the raw hyperspectral image through a cross-representation learning method. And then to learn semantically consistent representation over the created views by contrastive learning method. Specifically, four cross-channel-prediction based augmentation methods are naturally designed to utilize the high dimension characteristic of hyperspectral data for the view construction. And the better representative features are learned by maximizing mutual information and minimizing conditional entropy across different views from our contrastive network. This 'Cross-View-Predicton' style is straightforward and gets the state-of-the-art performance of unsupervised classification with a simple SVM classifier.