In e-commerce search, relevance between query and documents is an essential requirement for satisfying user experience. Different from traditional e-commerce platforms that offer products, users search on life service platforms such as Meituan mainly for product providers, which usually have abundant structured information, e.g. name, address, category, thousands of products. Modeling search relevance with these rich structured contents is challenging due to the following issues: (1) there is language distribution discrepancy among different fields of structured document, making it difficult to directly adopt off-the-shelf pretrained language model based methods like BERT. (2) different fields usually have different importance and their length vary greatly, making it difficult to extract document information helpful for relevance matching. To tackle these issues, in this paper we propose a novel two-stage pretraining and matching architecture for relevance matching with rich structured documents. At pretraining stage, we propose an effective pretraining method that employs both query and multiple fields of document as inputs, including an effective information compression method for lengthy fields. At relevance matching stage, a novel matching method is proposed by leveraging domain knowledge in search query to generate more effective document representations for relevance scoring. Extensive offline experiments and online A/B tests on millions of users verify that the proposed architectures effectively improve the performance of relevance modeling. The model has already been deployed online, serving the search traffic of Meituan for over a year.
Precise thigh muscle volumes are crucial to monitor the motor functionality of patients with diseases that may result in various degrees of thigh muscle loss. T1-weighted MRI is the default surrogate to obtain thigh muscle masks due to its contrast between muscle and fat signals. Deep learning approaches have recently been widely used to obtain these masks through segmentation. However, due to the insufficient amount of precise annotations, thigh muscle masks generated by deep learning approaches tend to misclassify intra-muscular fat (IMF) as muscle impacting the analysis of muscle volumetrics. As IMF is infiltrated inside the muscle, human annotations require expertise and time. Thus, precise muscle masks where IMF is excluded are limited in practice. To alleviate this, we propose a few-shot segmentation framework to generate thigh muscle masks excluding IMF. In our framework, we design a novel pseudo-label correction and evaluation scheme, together with a new noise robust loss for exploiting high certainty areas. The proposed framework only takes $1\%$ of the fine-annotated training dataset, and achieves comparable performance with fully supervised methods according to the experimental results.
Low earth orbit (LEO) satellite constellation-enabled communication networks are expected to be an important part of many Internet of Things (IoT) deployments due to their unique advantage of providing seamless global coverage. In this paper, we investigate the random access problem in massive multiple-input multiple-output-based LEO satellite systems, where the multi-satellite cooperative processing mechanism is considered. Specifically, at edge satellite nodes, we conceive a training sequence padded multi-carrier system to overcome the issue of imperfect synchronization, where the training sequence is utilized to detect the devices' activity and estimate their channels. Considering the inherent sparsity of terrestrial-satellite links and the sporadic traffic feature of IoT terminals, we utilize the orthogonal approximate message passing-multiple measurement vector algorithm to estimate the delay coefficients and user terminal activity. To further utilize the structure of the receive array, a two-dimensional estimation of signal parameters via rotational invariance technique is performed for enhancing channel estimation. Finally, at the central server node, we propose a majority voting scheme to enhance activity detection by aggregating backhaul information from multiple satellites. Moreover, multi-satellite cooperative linear data detection and multi-satellite cooperative Bayesian dequantization data detection are proposed to cope with perfect and quantized backhaul, respectively. Simulation results verify the effectiveness of our proposed schemes in terms of channel estimation, activity detection, and data detection for quasi-synchronous random access in satellite systems.
We consider the greedy algorithms for the joint recovery of high-dimensional sparse signals based on the block multiple measurement vector (BMMV) model in compressed sensing (CS). To this end, we first put forth two versions of simultaneous block orthogonal least squares (S-BOLS) as the baseline for the OLS framework. Their cornerstone is to sequentially check and select the support block to minimize the residual power. Then, parallel performance analysis for the existing simultaneous block orthogonal matching pursuit (S-BOMP) and the two proposed S-BOLS algorithms is developed. It indicates that under the conditions based on the mutual incoherence property (MIP) and the decaying magnitude structure of the nonzero blocks of the signal, the algorithms select all the significant blocks before possibly choosing incorrect ones. In addition, we further consider the problem of sufficient data volume for reliable recovery, and provide its MIP-based bounds in closed-form. These results together highlight the key role of the block characteristic in addressing the weak-sparse issue, i.e., the scenario where the overall sparsity is too large. The derived theoretical results are also universally valid for conventional block-greedy algorithms and non-block algorithms by setting the number of measurement vectors and the block length to 1, respectively.
The capacity of commercial massive multiple-input multiple-output (mMIMO) systems is constrained by the limited array aperture at the base station, and cannot meet the ever-increasing traffic demands of wireless networks. Given the array aperture, holographic MIMO with infinitesimal antenna spacing can maximize the capacity, but is physically unrealizable. As a promising alternative, reconfigurable mMIMO is proposed to harness the unexploited power of the electromagnetic (EM) domain for enhanced information transfer. Specifically, the reconfigurable pixel antenna technology provides each antenna with an adjustable EM radiation (EMR) pattern, introducing extra degrees of freedom for information transfer in the EM domain. In this article, we present the concept and benefits of availing the EMR domain for mMIMO transmission. Moreover, we propose a viable architecture for reconfigurable mMIMO systems, and the associated system model and downlink precoding are also discussed. In particular, a three-level precoding scheme is proposed, and simulation results verify its considerable spectral and energy efficiency advantages compared to traditional mMIMO systems. Finally, we further discuss the challenges, insights, and prospects of deploying reconfigurable mMIMO, along with the associated hardware, algorithms, and fundamental theory.
The dual-functional radar and communication (DFRC) technique constitutes a promising next-generation wireless solution, due to its benefits in terms of power consumption, physical hardware, and spectrum exploitation. In this paper, we propose sophisticated beamforming designs for multi-user DFRC systems by additionally taking the physical layer security (PLS) into account. We show that appropriately designed radar waveforms can also act as the traditional artificial noise conceived for drowning out the eavesdropping channel and for attaining increased design degrees of freedom (DoF). The joint beamforming design is formulated as a non-convex optimization problem for striking a compelling trade-off amongst the conflicting design objectives of radar transmit beampattern, communication quality of service (QoS), and the PLS level. Then, we propose a semidefinite relaxation (SDR)-based algorithm and a reduced-complexity version to tackle the non-convexity, where the globally optimal solutions are found. Moreover, a robust beamforming method is also developed for considering realistic imperfect channel state information (CSI) knowledge. Finally, simulation results are provided for corroborating our theoretical results and show the proposed methods' superiority.
As a rising star in the field of natural language processing, question answering systems (Q&A Systems) are widely used in all walks of life. Compared with other scenarios, the applicationin financial scenario has strong requirements in the traceability and interpretability of the Q&A systems. In addition, since the demand for artificial intelligence technology has gradually shifted from the initial computational intelligence to cognitive intelligence, this research mainly focuses on the financial numerical reasoning dataset - FinQA. In the shared task, the objective is to generate the reasoning program and the final answer according to the given financial report containing text and tables. We use the method based on DeBERTa pre-trained language model, with additional optimization methods including multi-model fusion, training set combination on this basis. We finally obtain an execution accuracy of 68.99 and a program accuracy of 64.53, ranking No. 4 in the 2022 FinQA Challenge.
Real-word search and recommender systems usually adopt a multi-stage ranking architecture, including matching, pre-ranking, ranking, and re-ranking. Previous works mainly focus on the ranking stage while very few focus on the pre-ranking stage. In this paper, we focus on the information transfer from ranking to pre-ranking stage. We propose a new Contrastive Information Transfer (CIT) framework to transfer useful information from ranking model to pre-ranking model. We train the pre-ranking model to distinguish the positive pair of representation from a set of positive and negative pairs with a contrastive objective. As a consequence, the pre-ranking model can make full use of rich information in ranking model's representations. The CIT framework also has the advantage of alleviating selection bias and improving the performance of recall metrics, which is crucial for pre-ranking models. We conduct extensive experiments including offline datasets and online A/B testing. Experimental results show that CIT achieves superior results than competitive models. In addition, a strict online A/B testing at one of the world's largest E-commercial platforms shows that the proposed model achieves 0.63\% improvements on CTR and 1.64\% improvements on VBR. The proposed model now has been deployed online and serves the main traffic of this system, contributing a remarkable business growth.
Rich user behavior data has been proven to be of great value for Click-Through Rate (CTR) prediction applications, especially in industrial recommender, search, or advertising systems. However, it's non-trivial for real-world systems to make full use of long-term user behaviors due to the strict requirements of online serving time. Most previous works adopt the retrieval-based strategy, where a small number of user behaviors are retrieved first for subsequent attention. However, the retrieval-based methods are sub-optimal and would cause more or less information losses, and it's difficult to balance the effectiveness and efficiency of the retrieval algorithm. In this paper, we propose \textbf{SDIM} (\textbf{S}ampling-based \textbf{D}eep \textbf{I}nterest \textbf{M}odeling), a simple yet effective sampling-based end-to-end approach for modeling long-term user behaviors. We sample from multiple hash functions to generate hash signatures of the candidate item and each item in the user behavior sequence, and obtain the user interest by directly gathering behavior items associated with the candidate item with the same hash signature. We show theoretically and experimentally that the proposed method performs on par with standard attention-based models on modeling long-term user behaviors, while being sizable times faster. We also introduce the deployment of SDIM in our system. Specifically, we decouple the behavior sequence hashing, which is the most time-consuming part, from the CTR model by designing a separate module named BSE (behavior Sequence Encoding). BSE is latency-free for the CTR server, enabling us to model extremely long user behaviors. Both offline and online experiments are conducted to demonstrate the effectiveness of SDIM. SDIM now has been deployed online in the search system of Meituan APP.