Yuanfeng Song

Generating executable structured queries for OpenStreetMap (OSM) from natural language remains a challenging task due to the rigid syntax of OverpassQL and the steep learning curve it presents to end-users. While steering Large Language Models (LLMs) via demonstration contexts offers a promising solution, existing retrieval-based approaches typically neglect the critical constraint of token budgets. This oversight often leads to prohibitive inference costs and ignores the combinatorial nature of context selection, resulting in suboptimal performance. Furthermore, the lack of reliable benchmarks hinders rigorous evaluation in this domain. To address these challenges, we first introduce OsmNL , a robust benchmark curated via a human-in-the-loop pipeline to resolve alignment noise in prior datasets. Building upon this foundation, we formulate the construction of optimal demonstration contexts under constraints as a Budgeted Example Subset Mining ( BESM ) problem. We analyze BESM as an NP-hard optimization task that admits a submodular objective, theoretically balancing semantic relevance with structural diversity. Leveraging this property, we propose TACO ( T oken- A ware C ontext O ptimization), a principled framework that instantiates a family of budget-aware algorithms to navigate the trade-off between approximation guarantees and computational efficiency. Extensive experiments on OsmNL show that TACO establishes a new state-of-the-art, achieving superior accuracy compared to strong baselines while reducing token consumption by over 50%, thereby offering a robust and cost-efficient solution for large-scale geospatial query interfaces.

In the era of big data, effectively leveraging vast amounts of data is one of the major challenges across industries, especially in data visualization, which helps decision-makers intuitively understand data. However, many existing data visualization tools require users to have certain technical skills, posing a barrier for users without a technical background. With the widespread use of voice interfaces, particularly smart assistants, the task of converting voice queries into data visualizations, known as ''Speech-to-Vis,'' has emerged. Yet, existing methods still face issues such as speech recognition errors and insufficient model generalization. We propose SD-DVQ, a Semantic-Driven Data Visualization Query Generation from Speech Queries framework, to address these issues. Unlike previous methods that rely on ASR-based cascades or completely end-to-end systems, SD-DVQ generates semantically consistent natural language queries from speech with schema information, and then improves downstream DVQ generation through schema filtering and database-feedback-based self-correction. Experiments on the SpeechNVBench benchmark show that SD-DVQ achieves 45.02% exact-match accuracy, outperforming the strongest baseline by 5.07 percentage points.

Recent advances in LLM-based agent systems enable natural language–driven, cross-modal data visualization. Yet, existing systems remain unreliable for complex, multi-modal, or iterative tasks. This demonstration introduces MultiVis-Agent, an LLM agent-based system enhanced with logic rules for reliable visualization generation. It integrates specialized agents for data querying, visualization, and validation under a central coordinator governed by a four-layer logic rule framework ensuring parameter safety, error recovery, and guaranteed termination. Built upon the MultiVis task—covering four scenarios from basic generation to iterative refinement—MultiVis-Agent produces executable visualization code from natural language, databases, and optional references. We will show that MultiVis-Agent achieves real-time generation, execution, and refinement with 99% task completion and strong cross-modal reasoning. In particular, we emphasize an end-to-end and interactive demonstration where users can upload databases, provide multi-modal references (images/code), and iteratively refine executable visualization code within the same interface. The demo video and code are available at https://github.com/Jinwei-Lu/MultiVis.

Real-world visualization tasks involve complex, multi-modal requirements that extend beyond simple text-to-chart generation, requiring reference images, code examples, and iterative refinement. Current systems exhibit fundamental limitations: single-modality input, one-shot generation, and rigid workflows. While LLM-based approaches show potential for these complex requirements, they introduce reliability challenges including catastrophic failures and infinite loop susceptibility. To address this gap, we propose MultiVis-Agent, a logic rule-enhanced multi-agent framework for reliable multi-modal and multi-scenario visualization generation. Our approach introduces a four-layer logic rule framework that provides mathematical guarantees for system reliability while maintaining flexibility. Unlike traditional rule-based systems, our logic rules are mathematical constraints that guide LLM reasoning rather than replacing it. We formalize the MultiVis task spanning four scenarios from basic generation to iterative refinement, and develop MultiVis-Bench, a benchmark with over 1,000 cases for multi-modal visualization evaluation. Extensive experiments demonstrate that our approach achieves 75.63% visualization score on challenging tasks, significantly outperforming baselines (57.54-62.79%), with task completion rates of 99.58% and code execution success rates of 94.56% (vs. 74.48% and 65.10% without logic rules), successfully addressing both complexity and reliability challenges in automated visualization generation.

Automated medication recommendation is a crucial task within the domain of artificial intelligence in healthcare, where recommender systems are supposed to deliver precise, personalized drug combinations tailored to the evolving health states of patients. Existing approaches often treat clinical records (e.g., diagnoses, procedures) as isolated or unified entities, neglecting the inherent set-structured nature of medical data and the need to model interdependencies among clinical elements. To address the gap, we propose SSPNet, a novel end-to-end framework designed to process complete clinical record sets and directly generate optimal medication sets. SSPNet employs a set-based encoder to effectively capture and represent a patient's health condition from the electronic health records (EHRs), while a permutation-consistent decoder predicts the entire medication combination as a set. In addition, we introduce a novel personalized representation mechanism to capture the drugs previously used by individual patients. Extensive experiments on MIMIC-Ⅲ and MIMIC-Ⅳ data sets reveal that SSPNet surpasses existing state-of-the-art methods in the accuracy of medication recommendations.