The rapid growth of e-commerce has driven an increasing demand for robotized warehouses to handle large-scale logistics orders. Upon receiving orders, a warehouse engages in task planning that involves two crucial stages: matching the orders with racks that contain the required items, and planning the robot routes to deliver those racks for order fulfillment. Hence, effective task planning is essential for maximizing order throughput. However, while existing techniques perform well for orders that involve items from a single rack, they exhibit low efficiency and poor performance when dealing with complex orders that require multiple items from different racks. In this paper, we introduce the robotized warehouse complex task planning problem and propose a novel Complex Order Online Planning (COOP) framework to address the challenge. Specifically, the framework matches orders with racks using a maximal coverage matching method, optimized through vector similarity search and a residual matching strategy. Then, it adopts an effective progressive prioritized pathfinding algorithm to transport matched racks with minimal delivery cost. Finally, the framework introduces an enhanced pathfinding-aware rack selection model that considers rack delivery costs from the pathfinding stage to collaboratively optimize rack matching and overall planning scheme. Extensive experiments on real-world and synthetic datasets demonstrate that our approaches exhibit strong performance across various parameter configurations.

With the proliferation of mobile devices, spatial crowdsourcing has emerged as a promising paradigm for facilitating location-based services, encompassing various applications across academia and industries. Recently, pioneering works have attempted to infer workers' mobility patterns from historical data to improve the quality of task assignment. However, these studies have overlooked or under-examined issues such as the dynamic mobility patterns of crowd workers, especially in the context of newcomers, the misalignment between the objectives of mobility prediction and task assignment, and the effective utilization of predicted mobility patterns. In this paper, we investigate a problem we term Task Assignment in Mobility Prediction-aware Spatial Crowdsourcing (TAMP). To address the TAMP problem, we first propose a task-adaptive meta-learning algorithm, which trains a set of specific meta-knowledge for workers' mobility prediction models through game theory-based learning task clustering and meta-training within each cluster. Then, we design a task assignment-oriented loss function and develop a task assignment algorithm that incorporates prediction performance, prioritizing assignments with higher confidence of completion. Extensive experiments on real-world datasets validate that our proposed methods can effectively improve the quality of task assignment.

Business intelligence (BI) transforms large volumes of data within modern organizations into actionable insights for informed decision-making. Recently, large language model (LLM)-based agents have streamlined the BI workflow by automatically performing task planning, reasoning, and actions in executable environments based on natural language (NL) queries. However, existing approaches primarily focus on individual BI tasks such as NL2SQL and NL2VIS. The fragmentation of tasks across different data roles and tools lead to inefficiencies and potential errors due to the iterative and collaborative nature of BI. In this paper, we introduce DataLab, a unified BI platform that integrates a one-stop LLM-based agent framework with an augmented computational notebook interface. DataLab supports various BI tasks for different data roles in data preparation, analysis, and visualization by seamlessly combining LLM assistance with user customization within a single environment. To achieve this unification, we design a domain knowledge incorporation module tailored for enterprise-specific BI tasks, an inter-agent communication mechanism to facilitate information sharing across the BI workflow, and a cell-based context management strategy to enhance context utilization efficiency in BI notebooks. Extensive experiments demonstrate that DataLab achieves state-of-the-art performance on various BI tasks across popular research benchmarks. Moreover, DataLab maintains high effectiveness and efficiency on real-world datasets from Tencent, achieving up to a 58.58% increase in accuracy and a 61.65 % reduction in token cost on enterprise-specific BI tasks.

Machine learning (ML) has become a cornerstone of key applications at ByteDance. As model complexity and data volumes surge, data management for large-scale ML workloads faces substantial challenges, particularly with recent advances in large recommendation models (LRMs) and large multimodal models (LMMs). Traditional approaches exhibit limitations in storage efficiency, metadata scalability, update mechanisms, and integration with ML frameworks. To address these challenges, we propose Magnus, a holistic data management system built upon Apache Iceberg. Magnus integrates innovative optimizations across resource-efficient storage formats optimized for large wide tables and multimodal data, built-in support for vector and inverted indexes to accelerate data retrieval, scalable metadata planning with Git-like branching and tagging capabilities, and high-performance update/upsert based on lightweight merge-on-read (MOR) strategies. Additionally, Magnus provides native support and specialized enhancement for LRM and LMM training workloads. Experimental results demonstrate significant performance gains in real-world ML scenarios. Magnus has been deployed at ByteDance for over five years, enabling robust and efficient data infrastructure for large-scale ML workloads.

In causal inference, estimating heterogeneous treatment effects (HTE) is critical for identifying how different subgroups respond to interventions, with broad applications in fields such as precision medicine and personalized advertising. Although HTE estimation methods aim to improve accuracy, how to provide explicit subgroup descriptions remains unclear, hindering data interpretation and strategic intervention management. In this paper, we propose CURLS, a novel rule learning method leveraging HTE, which can effectively describe subgroups with significant treatment effects. Specifically, we frame causal rule learning as a discrete optimization problem, finely balancing treatment effect with variance and considering the rule interpretability. We design an iterative procedure based on the minorize-maximization algorithm and solve a submodular lower bound as an approximation for the original. Quantitative experiments and qualitative case studies verify that compared with state-of-the-art methods, CURLS can find subgroups where the estimated and true effects are 16.1% and 13.8% higher and the variance is 12.0% smaller, while maintaining similar or better estimation accuracy and rule interpretability. Code is available at https://osf.io/zwp2k/.

Top- k Publish/Subscribe (TkPS) service is widely studied in spatial database, with various cache-based methods proposed to address its efficiency challenge in top- k result maintenance. These methods require in-depth exploration of relationships between cache updates and different factors (e.g., data distribution) to optimize cache performance. However, there is currently no system available that assists developers in conducting comprehensive cache analyses within TkPS services. We therefore introduce Catcher , a multi-functional cache analysis system designed for TkPS services. It not only enables users to intuitively analyze the entire maintenance process of top- k results but also aids in identifying bottlenecks and potential optimization spaces of caches. Catcher provides two user-friendly interfaces that allow users to employ simple and easy-to-use consoles to perform statistical analysis. Furthermore, Catcher offers the real-time evaluation of cache-based methods, providing users with instant analysis. We have demonstrated the usability of Catcher on real-world datasets. A short video of our demonstration can be found at https://youtu.be/qI81HoypB0w.

Patterns in charts refer to interesting visual features or forms. Identifying patterns not only helps analysts understand the ‘shape’ of the data but also supports better and faster decision-making. Existing solutions for identifying patterns in charts require a large number of labeled data instances, making it intractable without user supervision. In this paper, we propose ChartNavigator, an interactive pattern identification and annotation framework for unlabeled visualization charts. ChartNavigator leverages a novel chart-sensitive deep factor model to map patterns into a low-dimensional factor representation space, and facilitates rich analysis with the derived representations. We design and implement a visual interface to support efficient identification and annotation of potential patterns in charts. Evaluations with multiple datasets show that our approach outperforms the baseline models in identifying and annotating patterns

Distributed processing and analysis of large-scale graph data remain challenging because of the high-level discrepancy among graphs. This study investigates a novel subproblem: the distributed multi-task learning on the graph, which jointly learns multiple analysis tasks from decentralized graphs. We propose a federated multi-task graph learning (FMTGL) framework to solve the problem within a privacy-preserving and scalable scheme. Its core is an innovative data-fusion mechanism and a low-latency distributed optimization method. The former captures multi-source data relatedness and generates universal task representation for local task analysis. The latter enables the quick update of our framework with gradients sparsification and tree-based aggregation. As a theoretical result, the proposed optimization method has a convergence rate interpolates between \( \mathcal {O}(1/T) \) and \( \mathcal {O}(1/\sqrt {T}) \) , up to logarithmic terms. Unlike previous studies, our work analyzes the convergence behavior with adaptive stepsize selection and non-convex assumption. Experimental results on three graph datasets verify the effectiveness and scalability of FMTGL.

The increased accessibility of urban sensor data and the popularity of social network applications is enabling the discovery of crowd mobility and personal communication patterns. However, studying the egocentric relationships of an individual can be very challenging because available data may refer to direct contacts, such as phone calls between individuals, or indirect contacts, such as paired location presence. In this article, we develop methods to integrate three facets extracted from heterogeneous urban data (timelines, calls, and locations) through a progressive visual reasoning and inspection scheme. Our approach uses a detect-and-filter scheme such that, prior to visual refinement and analysis, a coarse detection is performed to extract the target individual and construct the timeline of the target. It then detects spatio-temporal co-occurrences or call-based contacts to develop the egocentric network of the individual. The filtering stage is enhanced with a line-based visual reasoning interface that facilitates a flexible and comprehensive investigation of egocentric relationships and connections in terms of time, space, and social networks. The integrated system, RelationLines, is demonstrated using a dataset that contains taxi GPS data, cell-base mobility data, mobile calling data, microblog data, and point-of-interest (POI) data from a city with millions of citizens. We examine the effectiveness and efficiency of our system with three case studies and user review.

Even with simple everyday tasks like online shopping or choosing a restaurant, users are easily overwhelmed with the large number of choices available today, each with a large number of inter-related attributes. We present ExRank, an interactive interface for exploring data that helps users understand the relationship between attribute values and find interesting items in the dataset. Based on a kNN graph and a PageRank algorithm, ExRank suggests which attributes the user should look at, and how expressed choices in particular attributes affect the distribution of values in other attributes for candidate objects. It solves the problem of empty result by showing similar items and when there are too many results, it ranks the data for the user. This demo consists of 1) the description of the software architecture and the user interface 2) the logic and reason behind our solution and 3) a list of demonstration scenarios for showing to the audience.

The popularity of social media services has been innovating the way of information acquisition in modern society. Meanwhile, mass information is generated in every single day. To extract useful knowledge, much effort has been invested in analyzing social media contents, e.g., (emerging) topic discovery. With these findings, however, users may still find it hard to obtain knowledge of great interest in conformity with their preference. In this paper, we present a novel system which brings proper context to continuously incoming social media contents, such that mass information can be indexed, organized and analyzed around Wikipedia entities. Four data analytics tools are employed in the system. Three of them aim to enrich each Wikipedia entity by analyzing the relevant contents while the other one builds an information network among the most relevant Wikipedia entities. With our system, users can easily pinpoint valuable information and knowledge they are interested in, as well as navigate to other closely related entities through the information network for further exploration.

Social data refers to data individuals create that is knowingly and voluntarily shared by them and is an exciting avenue into gaining insight into interpersonal behaviors and interaction. However, such data is large, heterogeneous and often incomplete, properties that make the analysis of such data extremely challenging. One common method of exploring such data is through cluster analysis, which can enable analysts to find groups of related users, behaviors and interactions. This paper presents a novel visual analysis approach for detecting clusters within large-scale social networks by utilizing a divide-analyze-recombine scheme that sequentially performs data partitioning, subset clustering and result recombination within an integrated visual interface. A case study on a microblog messaging data (with 4.8 millions users) is used to demonstrate the feasibility of this approach and comparisons are also provided to illustrate the performance benefits of this approach with respect to existing solutions.