Predicting a user's next search query from recent interaction behaviors is a critical problem in modern e-commerce systems, particularly in scenarios where user intent evolves rapidly. Large Language Models (LLMs) offer strong semantic reasoning capabilities and have recently been adopted to enhance training data construction for next-query prediction. However, due to resource constraints on mobile devices, existing applications are deployed on cloud servers, resulting in high inference costs. In this paper, we propose RecGPT-Mobile, a framework that designs a lightweight LLM-based intent understanding agent to improve recommendation quality in mobile e-commerce scenarios. By deploying LLM directly on mobile devices, our approach can capture the evolving interests of users more quickly and adjust the recommendation results in real time. Extensive offline analyzes and online experiments demonstrate that our method significantly improves the accuracy of recommendation results, laying a practical path for LLM deployment in production-scale recommendation systems on mobile devices, as well as a scalable solution for integrating LLMs into real-world next-query prediction systems.

A hierarchical labeling system for mobile applications (apps) benefits a wide range of downstream businesses that integrate the labeling with their proprietary user data, to improve user modeling. Such a label hierarchy can define more granular labels that capture detailed app features beyond the limitations of traditional broad app categories. In this paper, we address the problem of hierarchical multilabel classification for apps by using their textual information such as names and descriptions. We present: 1) HMCN (Hierarchical Multilabel Classification Network) for handling the classification from two perspectives: the first focuses on a multilabel classification without hierarchical constraints, while the second predicts labels sequentially at each hierarchical level considering such constraints; 2) HMCL (Hierarchical Multilabel Contrastive Learning), a scheme that is capable of learning more distinguishable app representations to enhance the performance of HMCN. Empirical results on our Tencent App Store dataset and two public datasets demonstrate that our approach performs well compared with state-of-the-art methods. The approach has been deployed at Tencent and the multilabel classification outputs for apps have helped a downstream task--credit risk management of users--improve its performance by 10.70% with regard to the Kolmogorov-Smirnov metric, for over one year.

With the development of machine vision and multimedia technology, posture detection and related algorithms have become widely used in the field of human posture recognition. Traditional video surveillance methods have the disadvantages of slow detection speed, low accuracy, interference from occlusions, and poor real-time performance. This paper proposes a real-time pose detection algorithm based on deep learning, which can effectively perform real-time tracking and detection of single and multiple individuals in different indoor and outdoor environments and at different distances. First, a corresponding pose recognition dataset for complex scenes was created based on the YOLO network. Then, the OpenPose method was used to detect key points of the human body. Finally, the Kalman filter multi-object tracking method was used to predict the state of human targets within the occluded area. Real-time detection of human postures (sitting, stand up, standing, sit down, walking, fall down, and lying down) is achieved with corresponding alarms to ensure the timely detection and processing of emergencies.

Very recently, semi-factual explanations have emerged in Explainable AI (XAI) as a new and potentially important explanation strategy. Semi-factuals employ “Even if...” reasoning, as opposed to the “If only...” reasoning of counterfactuals. Counterfactuals inform users about what feature-differences lead to changes in an outcome (e.g., “ if only you asked for a lower loan, you would have been successful.”), whereas semi-factuals inform them about what feature-differences lead to the outcome remaining the same (e.g., “ Even if you asked for a lower loan, you would still have been unsuccessful”). Semi-factuals have the potential to be as important as their popular counterfactual siblings. However, the AI/ML and XAI communities have by and large struggled to imagine useful application-scenarios for semi-factuals. In this paper, we summarize recent work on semi-factual explanation and trace a roadmap for application-focused research in the area. We begin by outlining the main constraints identified for semi-factual optimization proposed in the literature, before summarizing the applications of semi-factuals proposed to-date. Then, we sketch several directions for future applications and research using semi-factuals. Finally, though semi-factuals are highly promising (especially with regard to algorithmic recourse), they have a potential for ethical misuse that we discuss in our conclusions.

When users receive either a positive or negative outcome from an automated system, Explainable AI (XAI) has almost exclusively focused on how to mutate negative outcomes into positive ones by crossing a decision boundary using counterfactuals (e.g., *"If you earn 2k more, we will accept your loan application"*). Here, we instead focus on positive outcomes, and take the novel step of using XAI to optimise them (e.g., *"Even if you wish to half your down-payment, we will still accept your loan application"*). Explanations such as these that employ "even if..." reasoning, and do not cross a decision boundary, are known as semifactuals. To instantiate semifactuals in this context, we introduce the concept of *Gain* (i.e., how much a user stands to benefit from the explanation), and consider the first causal formalisation of semifactuals. Tests on benchmark datasets show our algorithms are better at maximising gain compared to prior work, and that causality is important in the process. Most importantly however, a user study supports our main hypothesis by showing people find semifactual explanations more useful than counterfactuals when they receive the positive outcome of a loan acceptance.

Depth completion is a crucial computer vision task that aims to fill in missing or incomplete depth values in a depth map. In this paper, we propose SDC: Spatial Depth Completion for Outdoor Scenes. Our approach leverages a two-stage architecture with a spatial feature extractor (SFE) to utilize multi-scale features for accurate depth completion effectively. The proposed method incorporates attention mechanisms, including the Efficient Position Attention Module (EPAM) and Channel Attention Module (CAM), to adaptively fuse depth map features and improve the accuracy of depth completion. Additionally, the Pearson loss function is employed to further enhance the accuracy of the completed depth maps. Experimental results on the KITTI depth completion benchmark demonstrate that our method achieves comparable or better results than traditional depth completion methods while significantly reducing the number of parameters. The proposed SDC model shows great potential in practical applications of depth completion, with its ability to effectively fuse multiscale features and compact model size.

k-means algorithm is one of the most classical clustering methods, which has been widely and successfully used in signal processing. However, due to the thin-tailed property of the Gaussian distribution, k-means algorithm suffers from relatively poor performance on the dataset containing heavy-tailed data or outliers. Besides, standard k-means algorithm also has relatively weak stability, i.e. its results have a large variance, which reduces its credibility. In this paper, we propose a robust and stable k-means variant, dubbed the t-k-means, as well as its fast version to alleviate those problems. Theoretically, we derive the t-k-means and analyze its robustness and stability from the aspect of the loss function and the expression of the clustering center, respectively. Extensive experiments are also conducted, which verify the effectiveness and efficiency of the proposed method. The code for reproducing main results is available at https://github.com/THUYimingLi/t-k-means.

Hierarchical clustering has been shown to be valuable in many scenarios. Despite its usefulness to many situations, there is no agreed methodology on how to properly evaluate the hierarchies produced from different techniques, particularly in the case where ground-truth labels are unavailable. This motivates us to propose a framework for assessing the quality of hierarchical clustering allocations which covers the case of no ground-truth information. This measurement is useful, e.g., to assess the hierarchical structures used by online retailer websites to display their product catalogues. Our framework is one of the few attempts for the hierarchy evaluation from a decision theoretic perspective. We model the process as a bot searching stochastically for items in the hierarchy and establish a measure representing the degree to which the hierarchy supports this search. We employ Partially Observable Markov Decision Processes (POMDP) to model the uncertainty, the decision making, and the cognitive return for searchers in such a scenario.

The Soft Affiliation Graph model (S-AGM) is a Bayesian generative model of overlapping community structure in social networks. Inference on this model is challenging due to the complexity of both the underlying network structure and the presence of non-conjugacy in the model. Scalable MCMC on the model is possible through the use of Stochastic Gradient Riemannian Langevin Dynamics (SGRLD). In this paper, we develop a novel and scalable Stochastic Gradient Variational Inference (SG-VI) algorithm and compare it to SGRLD inference. Similarly to MCMC inference, handling non-conjugacy in the S-AGM is a significant challenge for developing an SG-VI and requires the application of stochastic Monte Carlo estimation. We carry out a thorough empirical comparison of the SG-VI and SGRLD approaches, and draw some general conclusions about scalable inference on the S-AGM.

The ambiguous annotation criteria lead to divergence of Chinese Word Segmentation (CWS) datasets in various granularities.Multi-criteria Chinese word segmentation aims to capture various annotation criteria among datasets and leverage their common underlying knowledge.In this paper, we propose a domain adaptive segmenter to exploit diverse criteria of various datasets.Our model is based on Bidirectional Encoder Representations from Transformers (BERT), which is responsible for introducing open-domain knowledge.Private and shared projection layers are proposed to capture domain-specific knowledge and common knowledge, respectively.We also optimize computational efficiency via distillation, quantization, and compiler optimization.Experiments show that our segmenter outperforms the previous state of the art (SOTA) models on 10 CWS datasets with superior efficiency.

Virtual Reality (VR) produces a highly realistic simulation environment to engage users with Immersive Virtual Environments (IVEs). To interact effectively with users, VR builds intensive media through the multi-modal sense functions in the lower level, such as visual, auditory, tactile, and olfactory senses. However, the higher-level perceptions, e.g., the temporal duration, the sense of presence, and the cognitive load are less explored. These higher-level perceptions are part of the critical evaluation criteria for VR design. In this paper, we divide the external zeitgebers into visual and auditory zeitgebers. We then combine these zeitgebers with the attention-oriented cognitive load to investigate their effects on temporal estimation and presence, particularly in IVEs. We propose a data-driven method to build a multi-modal predictive equation for time estimation and presence, in an effort to figure out the essential elements of users' spatial and temporal perception in VR. We also design a complicated application and validate the predictive equation. Our feature-based model is able to guide the VR application design in terms of the subjective time length judgment and presence of users as well as achieve a better VR user experience.

Aspect-term sentiment analysis (ATSA) is a long-standing challenge in natural language processing.It requires fine-grained semantical reasoning about a target entity appeared in the text.As manual annotation over the aspects is laborious and time-consuming, the amount of labeled data is limited for supervised learning.This paper proposes a semisupervised method for the ATSA problem by using the Variational Autoencoder based on Transformer.The model learns the latent distribution via variational inference.By disentangling the latent representation into the aspect-specific sentiment and the lexical context, our method induces the underlying sentiment prediction for the unlabeled data, which then benefits the ATSA classifier.Our method is classifier-agnostic, i.e., the classifier is an independent module and various supervised models can be integrated.Experimental results are obtained on the SemEval 2014 task 4 and show that our method is effective with different five specific classifiers and outperforms these models by a significant margin.

Person search is a challenging task that requires to address pedestrian detection and person re- identification simultaneously. Though significant progress has been made in detection and re-identification respectively, the similar appearances of persons, pedestrian misdetections and false alarms still have adverse effects on person search. To this end, an improved end-to-end person search network with multi -loss is proposed to jointly optimize detection and re-identification. Firstly, a pre-trained network is designed to obtain proper initial state for the whole training network. Then, to enhance the person search model, an improved online instance matching (IOIM) loss is proposed by hardening the distribution of labeled identities and softening the distribution of unlabeled identities. Finally, considering the intra-class compactness of features learned by center loss, the IOIM loss is combined with center loss by the proposed multi-loss fusion strategy, which can learn more discriminative feature embeddings. Experimental results on two challenging datasets CUHK-SYSU and PRW demonstrate our approach significantly outperforms the state-of-the-arts.

Loop closure detection is essential and important in visual simultaneous localization and mapping (SLAM) systems. Most existing methods typically utilize a separate feature extraction part and a similarity metric part. Compared to these methods, an end-to-end network is proposed in this paper to jointly optimize the two parts in a unified framework for further enhancing the interworking between these two parts. First, a two-branch siamese network is designed to learn respective features for each scene of an image pair. Then a hierarchical weighted distance (HWD) layer is proposed to fuse the multi-scale features of each convolutional module and calculate the distance between the image pair. Finally, by using the contrastive loss in the training process, the effective feature representation and similarity metric can be learned simultaneously. Experiments on several open datasets illustrate the superior performance of our approach and demonstrate that the end-to-end network is feasible to conduct the loop closure detection in real time and provides an implementable method for visual SLAM systems.

Person search, which is vital for intelligent surveillance, aims at detecting and re-identifying pedestrians from whole monitoring images. However, due to the inaccurate pedestrian detections and extremely few instances per training identity, it remains challenging to learn discriminative representations only by labeled identities for person search. To this end, this paper proposes a novel loss function called instance enhancing loss (IEL) to learn deep identity-sensitive features by introducing unlabeled identity information. Specifically, the proposed IEL can selectively annotate unlabeled identities with similar appearances to labeled identities, and utilize these unlabeled identities in conjunction with labeled identities to train the person search network. The amount of unlabeled identities used as labeled instances can be quantitatively adjusted. Moreover, the proposed IEL is trainable and easy to optimize by back propagation algorithms. Extensive experiments on two benchmark datasets, namely CUHK-SYSU and PRW, show that our method outperforms state-of-the-arts for person search.

Person re-identification remains a challenging problem due to large variations of poses, occlusions, illumination and camera views. To learn both feature representation and similarity metric simultaneously, deep metric learning methods using triplet convolutional neural network have been applied in person re-identification. In this paper, we propose a body structure based triplet convolutional neural network (BSTCNN) for person re-identification. Specifically, a four-branch CNN architecture is built to learn features from different body parts. Body-part features are then fused in score level with a novel weighted distance layer which learns weights for each body part. We further design an improved triplet loss function to speed up convergence and boost the performance. Experimental results on two challenging datasets (CUHK01 and PRID2011) demonstrate that our approach significantly outperforms the state-of-the-art methods.

In epidemic modeling, state filtering is an excellent tool for enhancing the performance of traditional epidemic models. We introduce a novel state filter algorithm to further improve the performance of state-of-the-art approaches based on Susceptible-Infected-Recovered (SIR) models. The proposed algorithm merges two techniques, which are typically used separately: linear correction, as seen in the Ensemble Kalman Filter (EnKF), and resampling, as used in the Particle Filter (PF). We compare the inferential accuracy of our approach against the EnKF and the Ensemble Adjustment Kalman Filter (EAKF), using algorithms employing both an uncentered covariance matrix (UCM) and the standard column-centered covariance matrix (CCM). Our algorithm requires O(DN) more time than EnKF does, where D is the ensemble dimension and N denotes the ensemble size. We demonstrate empirically that our algorithm with UCM achieves the lowest root-mean-square-error (RMSE) and the highest correlation coefficient (CORR) amongst the selected methods, in 11 out of 14 major real-world scenarios. We show that the EnKF with UCM outperforms the EnKF with CCM, while the EAKF gains better accuracy with CCM in most scenarios.

The effectiveness of wireless sensor networks (WSN) depends on the coverage and connectivity provided by node deployment, which is one of the key topics in WSN. In this paper, a modified virtual force-based node self-deployment algorithm for nodes with mobility is proposed. In the virtual force-based approach, all nodes are seen as points subject to repulsive and attractive force exerted among them, nodes can move according to the calculated force. In the proposed approach, Delaunay triangulation is formed with these nodes, adjacent relationship is defined if two nodes are connected in the Delaunay diagram. Force can only be exerted from those adjacent nodes within the communication range. Simulation results showed that the proposed approach has higher coverage rate and faster convergence time than traditional virtual force algorithm.

The effectiveness of wireless sensor networks(WSN) depends on the coverage and connectivity provided by node deployment, which is one of the key topics in WSN. In this paper, a virtual force-based self-deployment algorithm for nodes with mobility using a force model similar to van der Waals force is proposed. In the proposed approach, all nodes are seen as points subject to repulsive and attractive force exerted among them, nodes can move according to the calculated force. Simulation results and comparisons have showed that the proposed approach has higher coverage rate, faster convergence time and less energy consuming than traditional virtual force algorithm.