In this paper, we contribute a novel and extensive dataset for speaker verification, which contains noisy 38k identities/1.45M utterances (VoxBlink) and relatively cleaned 18k identities/1.02M (VoxBlink-Clean) utterances for training. Firstly, we collect a 60K+ users' list as well as their avatar and download their SHORT videos on the YouTube. Then, an automatically pipeline is devised to extract target user's speech segments and videos, which is efficient and scalable. To the best of our knowledge, the VoxBlink dataset is the largest speaker recognition dataset. Secondly, we develop a series of experiments based on VoxBlink-clean together with VoxCeleb2. Our findings highlight a notable improvement in performance, ranging from 15% to 30%, across different backbone architectures, upon integrating our dataset for training. The dataset will be released SOON~.
In this paper, we contribute a novel and extensive dataset for speaker verification, which contains noisy 38k identities/1.45M utterances (VoxSnap) and relatively cleaned 18k identities/1.02M (VoxSnap-Clean) utterances for training. Firstly, we collect a 60K+ users' list as well as their avatar and download their SHORT videos on the YouTube. Then, an automatically pipeline is devised to extract target user's speech segments and videos, which is efficient and scalable. To the best of our knowledge, the VoxSnap dataset is the largest speaker recognition dataset. Secondly, we develop a series of experiments based on VoxSnap-clean together with VoxCeleb2. Our findings highlight a notable improvement in performance, ranging from 15% to 30%, across different backbone architectures, upon integrating our dataset for training. The dataset will be released SOON~.
In this work, we conceptualize the learning process as information compression. We seek to equip generative pre-trained models with human-like learning capabilities that enable data compression during inference. We present a novel approach that utilizes the Generative Pre-trained Transformer (GPT) to approximate Kolmogorov complexity, with the aim of estimating the optimal Information Distance for few-shot learning. We first propose using GPT as a prior for lossless text compression, achieving a noteworthy compression ratio. Experiment with LLAMA2-7B backbone achieves a compression ratio of 15.5 on enwik9. We justify the pre-training objective of GPT models by demonstrating its equivalence to the compression length, and, consequently, its ability to approximate the information distance for texts. Leveraging the approximated information distance, our method allows the direct application of GPT models in quantitative text similarity measurements. Experiment results show that our method overall achieves superior performance compared to embedding and prompt baselines on challenging NLP tasks, including semantic similarity, zero and one-shot text classification, and zero-shot text ranking.
Integrated sensing and communications (ISAC) is emerging as a key technique for next-generation wireless systems. In order to enable ubiquitous mobile network and widely deployed base stations with radar sensing functionality, using standard multiple-input multiple-output (MIMO) orthogonal frequency division multiplexing (OFDM) transmitter architectures and waveforms to achieve both satisfactory sensing and communication performance is a crucial task for practical implementations of ISAC. In this paper, we propose a novel joint angle-range-velocity estimation algorithm for the considered MIMO-OFDM ISAC system. The approach is based on classical MIMO-OFDM waveforms widely adopted in wireless communications. Specifically, the angle-range-velocity information of potential targets is jointly extracted by utilizing all the received echo signals in a coherent processing interval (CPI). Therefore, the proposed joint estimation algorithm can achieve larger processing gains and higher resolution by fully exploiting echo signals and jointly estimating the angle-range-velocity information. Theoretical analysis for maximum unambiguous range, resolution, and processing gains are provided to verify the advantages of the proposed joint estimation algorithm. Finally, extensive numerical experiments are presented to demonstrate that the proposed estimation approach can achieve significantly lower root-mean-square-error (RMSE) of angle/range/velocity estimation for both single-target and multi-target scenarios.
Next basket recommendation (NBR) is the task of predicting the next set of items based on a sequence of already purchased baskets. It is a recommendation task that has been widely studied, especially in the context of grocery shopping. In next basket recommendation (NBR), it is useful to distinguish between repeat items, i.e., items that a user has consumed before, and explore items, i.e., items that a user has not consumed before. Most NBR work either ignores this distinction or focuses on repeat items. We formulate the next novel basket recommendation (NNBR) task, i.e., the task of recommending a basket that only consists of novel items, which is valuable for both real-world application and NBR evaluation. We evaluate how existing NBR methods perform on the NNBR task and find that, so far, limited progress has been made w.r.t. the NNBR task. To address the NNBR task, we propose a simple bi-directional transformer basket recommendation model (BTBR), which is focused on directly modeling item-to-item correlations within and across baskets instead of learning complex basket representations. To properly train BTBR, we propose and investigate several masking strategies and training objectives: (i) item-level random masking, (ii) item-level select masking, (iii) basket-level all masking, (iv) basket-level explore masking, and (v) joint masking. In addition, an item-basket swapping strategy is proposed to enrich the item interactions within the same baskets. We conduct extensive experiments on three open datasets with various characteristics. The results demonstrate the effectiveness of BTBR and our masking and swapping strategies for the NNBR task. BTBR with a properly selected masking and swapping strategy can substantially improve NNBR performance.
The paradigm of large-scale pre-training followed by downstream fine-tuning has been widely employed in various object detection algorithms. In this paper, we reveal discrepancies in data, model, and task between the pre-training and fine-tuning procedure in existing practices, which implicitly limit the detector's performance, generalization ability, and convergence speed. To this end, we propose AlignDet, a unified pre-training framework that can be adapted to various existing detectors to alleviate the discrepancies. AlignDet decouples the pre-training process into two stages, i.e., image-domain and box-domain pre-training. The image-domain pre-training optimizes the detection backbone to capture holistic visual abstraction, and box-domain pre-training learns instance-level semantics and task-aware concepts to initialize the parts out of the backbone. By incorporating the self-supervised pre-trained backbones, we can pre-train all modules for various detectors in an unsupervised paradigm. As depicted in Figure 1, extensive experiments demonstrate that AlignDet can achieve significant improvements across diverse protocols, such as detection algorithm, model backbone, data setting, and training schedule. For example, AlignDet improves FCOS by 5.3 mAP, RetinaNet by 2.1 mAP, Faster R-CNN by 3.3 mAP, and DETR by 2.3 mAP under fewer epochs.
To improve the robustness of graph neural networks (GNN), graph structure learning (GSL) has attracted great interest due to the pervasiveness of noise in graph data. Many approaches have been proposed for GSL to jointly learn a clean graph structure and corresponding representations. To extend the previous work, this paper proposes a novel regularized GSL approach, particularly with an alignment of feature information and graph information, which is motivated mainly by our derived lower bound of node-level Rademacher complexity for GNNs. Additionally, our proposed approach incorporates sparse dimensional reduction to leverage low-dimensional node features that are relevant to the graph structure. To evaluate the effectiveness of our approach, we conduct experiments on real-world graphs. The results demonstrate that our proposed GSL method outperforms several competitive baselines, especially in scenarios where the graph structures are heavily affected by noise. Overall, our research highlights the importance of integrating feature and graph information alignment in GSL, as inspired by our derived theoretical result, and showcases the superiority of our approach in handling noisy graph structures through comprehensive experiments on real-world datasets.
This paper introduces the Multi-scale Feature Aggregation Conformer (MFA-Conformer) structure for audio anti-spoofing countermeasure (CM). MFA-Conformer combines a convolutional neural networkbased on the Transformer, allowing it to aggregate global andlocal information. This may benefit the anti-spoofing CM system to capture the synthetic artifacts hidden both locally and globally. In addition, given the excellent performance of MFA Conformer on automatic speech recognition (ASR) and automatic speaker verification (ASV) tasks, we present a transfer learning method that utilizes pretrained Conformer models on ASR or ASV tasks to enhance the robustness of CM systems. The proposed method is evaluated on both Chinese and Englishs poofing detection databases. On the FAD clean set, the MFA-Conformer model pretrained on the ASR task achieves an EER of 0.038%, which dramatically outperforms the baseline. Moreover, experimental results demonstrate that proposed transfer learning method on Conformer is effective on pure speech segments after voice activity detection processing.
This paper investigates the problem of regret minimization for multi-armed bandit (MAB) problems with local differential privacy (LDP) guarantee. Given a fixed privacy budget $\epsilon$, we consider three privatizing mechanisms under Bernoulli scenario: linear, quadratic and exponential mechanisms. Under each mechanism, we derive stochastic regret bound for Thompson Sampling algorithm. Finally, we simulate to illustrate the convergence of different mechanisms under different privacy budgets.
This paper studies an intelligent reflecting surface (IRS)-aided multi-antenna simultaneous wireless information and power transfer (SWIPT) system where an $M$-antenna access point (AP) serves $K$ single-antenna information users (IUs) and $J$ single-antenna energy users (EUs) with the aid of an IRS with phase errors. We explicitly concentrate on overloaded scenarios where $K + J > M$ and $K \geq M$. Our goal is to maximize the minimum throughput among all the IUs by optimizing the allocation of resources (including time, transmit beamforming at the AP, and reflect beamforming at the IRS), while guaranteeing the minimum amount of harvested energy at each EU. Towards this goal, we propose two user grouping (UG) schemes, namely, the non-overlapping UG scheme and the overlapping UG scheme, where the difference lies in whether identical IUs can exist in multiple groups. Different IU groups are served in orthogonal time dimensions, while the IUs in the same group are served simultaneously with all the EUs via spatial multiplexing. The two problems corresponding to the two UG schemes are mixed-integer non-convex optimization problems and difficult to solve optimally. We propose efficient algorithms for these two problems based on the big-M formulation, the penalty method, the block coordinate descent, and the successive convex approximation. Simulation results show that: 1) the non-robust counterparts of the proposed robust designs are unsuitable for practical IRS-aided SWIPT systems with phase errors since the energy harvesting constraints cannot be satisfied; 2) the proposed UG strategies can significantly improve the max-min throughput over the benchmark schemes without UG or adopting random UG; 3) the overlapping UG scheme performs much better than its non-overlapping counterpart when the absolute difference between $K$ and $M$ is small and the EH constraints are not stringent.