The creation of systematic literature reviews (SLR) is critical for analyzing the landscape of a research field and guiding future research directions. However, retrieving and filtering the literature corpus for an SLR is highly time-consuming and requires extensive manual effort, as keyword-based searches in digital libraries often return numerous irrelevant publications. In this work, we propose a pipeline leveraging multiple large language models (LLMs), classifying papers based on descriptive prompts and deciding jointly using a consensus scheme. The entire process is human-supervised and interactively controlled via our open-source visual analytics web interface, LLMSurver, which enables real-time inspection and modification of model outputs. We evaluate our approach using ground-truth data from a recent SLR comprising 8323 candidate papers, benchmarking both open and commercial state-of-the-art LLMs from mid-2024 and fall 2025. Results demonstrate that our pipeline significantly reduces manual effort while achieving lower error rates than single human annotators. Furthermore, modern open-source models prove sufficient for this task, making the method accessible and cost-effective. Overall, our work demonstrates how responsible human-AI collaboration can accelerate and enhance systematic literature reviews within academic workflows.

The maritime sector is undergoing a disruptive technological change driven by three main factors: autonomy, decarbonization, and digital transformation. Addressing these factors necessitates a reassessment of inland vessel operations. This paper presents the design and development of a decision support system for ferry operations based on a shrinking-horizon optimal control framework. The problem formulation incorporates a mathematical model of the ferry's dynamics and environmental disturbances, specifically water currents and wind, which can significantly influence the dynamics. Real-world data and illustrative scenarios demonstrate the potential of the proposed system to effectively support ferry crews by providing real-time guidance. This enables enhanced operational efficiency while maintaining predefined maneuver durations. The findings suggest that optimal control applications hold substantial promise for advancing future ferry operations on inland waters. A video of the real-world ferry MS Insel Mainau operating on Lake Constance is available at: https://youtu.be/i1MjCdbEQyE

We propose an interactive methodology for generating counterfactual explanations for univariate time series data in classification tasks by leveraging 2D projections and decision boundary maps to tackle interpretability challenges. Our approach aims to enhance the transparency and understanding of deep learning models' decision processes. The application simplifies the time series data analysis by enabling users to interactively manipulate projected data points, providing intuitive insights through inverse projection techniques. By abstracting user interactions with the projected data points rather than the raw time series data, our method facilitates an intuitive generation of counterfactual explanations. This approach allows for a more straightforward exploration of univariate time series data, enabling users to manipulate data points to comprehend potential outcomes of hypothetical scenarios. We validate this method using the ECG5000 benchmark dataset, demonstrating significant improvements in interpretability and user understanding of time series classification. The results indicate a promising direction for enhancing explainable AI, with potential applications in various domains requiring transparent and interpretable deep learning models. Future work will explore the scalability of this method to multivariate time series data and its integration with other interpretability techniques.

Detecting and interpreting common patterns in relational data is crucial for understanding complex topological structures across various domains. These patterns, or network motifs, can often be detected algorithmically. However, visual inspection remains vital for exploring and discovering patterns. This paper focuses on presenting motifs within BioFabric network visualizations—a unique technique that opens opportunities for research on scaling to larger networks, design variations, and layout algorithms to better expose motifs. Our goal is to show how highlighting motifs can assist users in identifying and interpreting patterns in BioFabric visualizations. To this end, we leverage existing motif simplification techniques. We replace edges with glyphs representing fundamental motifs such as staircases, cliques, paths, and connector nodes. The results of our controlled experiment and usage scenarios demonstrate that motif simplification for BioFabric is useful for detecting and interpreting network patterns. Our participants were faster and more confident using the simplified view without sacrificing accuracy. The efficacy of our current motif simplification approach depends on which extant layout algorithm is used. We hope our promising findings on user performance will motivate future research on layout algorithms tailored to maximizing motif presentation. Our supplemental material is available at https://osf.io/f8s3g/?view_only=7e2df9109dfd4e6c85b89ed828320843.

Searching in highly homogeneous video domains is challenging, especially when relying solely on human memory. This difficulty arises because homogeneous content often requires domain-specific vocabulary to describe effectively, and a single text label might fit a substantial subset of the database. This paper investigates supplementing text queries with spatial location information to better address specific search intents—a strategy applicable when users possess strong visual memory of a target object, or have external knowledge of its position. To process this location information effectively, we formally define and evaluate static grid and dynamic segmentation strategies for video frame partitioning. Furthermore, we present a large-scale cognitive user study involving 220 participants, designed to simulate realistic memory constraints of Known-Item Search (KIS) tasks commonly found in interactive retrieval benchmarks. The study utilizes standard working memory interference techniques, comprising a target exposure phase, a distraction period, and subsequent query specification from memory. These queries are then evaluated against the proposed spatial retrieval models to determine how human spatial memory decay impacts retrieval effectiveness. Our results reveal that while dynamic segmentation models achieve the highest theoretical retrieval bounds under perfect conditions, their strict geometric boundaries are severely affected by human memory decay. Consequently, static overlapping grids demonstrate high robustness, outperforming dynamic models under more realistic, memory-driven search constraints. Finally, our experiments show that the observed spatial annotation perturbations can be modeled using a four-dimensional Kernel Density Estimation (KDE) method, enabling the simulation of realistic human memory decay on ideal bounding boxes.

PraK V4 is an interactive video search system built for retrieval competitions. It introduces five advances over the prior release: (1) spatial conjunction of localized queries, letting users specify AND constraints across separate regions of interest, evaluated in a user study; (2) semi-automated user-interface adaptation for AVS tasks using online learning; (3) single-video browsing through a redesigned player and within-video querying; and (4) a parallelized backend for lower latency, (5) an updated keyframe layout strategy.

Parametric projections let analysts embed new points in real time, but input variations from measurement noise or data drift can produce unpredictable shifts in the 2D layout. Whether and where a projection is locally stable remains largely unexamined. In this paper, we present a stability evaluation framework that probes parametric projections with Gaussian perturbations around selected anchor points and assesses how neighborhoods deform in the 2D embedding. Our approach combines quantitative measures of mean displacement, bias, and nearest-anchor assignment error with per-anchor visualizations of displacement vectors, local PCA ellipsoids, and Voronoi misassignment for detailed inspection. We demonstrate the framework's effectiveness on UMAP- and t-SNE-based neural projectors of varying network sizes and study the effect of Jacobian regularization as a gradient-based robustness strategy. We apply our framework to the MNIST and Fashion-MNIST datasets. The results show that our framework identifies unstable projection regions invisible to reconstruction error or neighborhood-preservation metrics.

Multimodal deep-learning models power interactive video retrieval by ranking keyframes in response to textual queries. Despite these advances, users must still browse ranked candidates manually to locate a target. Keyframe arrangement within the search grid highly affects browsing effectiveness and user efficiency, yet remains underexplored. We report a study with 49 participants evaluating seven keyframe layouts for the Visual Known-Item Search task. Beyond efficiency and accuracy, we relate browsing phenomena, such as overlooks, to layout characteristics. Our results show that a video-grouped layout is the most efficient, while a four-column, rank-preserving grid achieves the highest accuracy. Sorted grids reveal potentials and trade-offs, enabling rapid scanning of uninteresting regions but down-ranking relevant targets to less prominent positions, delaying first arrival times and increasing overlooks. These findings motivate hybrid designs that preserve positions of top-ranked items while sorting or grouping the remainder, and offer guidance for searching in grids beyond video retrieval.

The analysis of spatiotemporal data is essential in domains such as epidemiology and environmental monitoring, where understanding the interplay between spatially distributed phenomena and their temporal evolution is critical. Dense pixel visualizations offer a compact, effective overview of spatiotemporal dynamics. However, the necessary linearization of 2D geographic space into a 1D ordering inevitably introduces structural distortions that manifest as visual artifacts. We propose a measure-driven visual analytics approach that captures visual artifacts through neighborhood preservation measures for 1D orderings and renders them using visual boosting techniques such as glyphs, halos, and hatching. We demonstrate our approach through a usage scenario analyzing COVID-19 incidence data across German districts, showing that interactive, measure-driven boosting enables analysts to reliably distinguish genuine spatial patterns from linearization artifacts.

Searching large video collections is typically an interactive process in which users play two roles. First, they hold the search intent: the underlying goal that determines what content they seek and why. Second, users must operationalize this intent through an iterative search loop. Users translate their intent into queries, browse the retrieved candidates, and refine their queries based on the results. In this paper, we investigate the capabilities of modern Vision Language Models (VLM) and agentic approaches to reach search goals interactively and fully autonomously. Specifically, we study whether a provided initial specification of a search goal might be sufficient to solve traditionally interactive search tasks with an agentic system. Provided that the involved VLMs are not aware of the whole large video dataset in advance, the key challenge lies in the effective combination of an existing interactive video search system and a smart VLM agent controlling the system. While the search system provides indexing and efficient querying, the VLM-based agents analyze top-ranked items and make decisions about next actions. Our results show that modern agents can autonomously operate interactive video retrieval systems to solve many search tasks from an initial intent description, achieving performance competitive with strong historical expert-operated systems in several settings.

Recently, neural networks have gained attention for creating parametric and invertible multidimensional data projections. Parametric projections allow for embedding previously unseen data without recomputing the projection as a whole, while invertible projections enable the generation of new data points. However, these properties have never been explored simultaneously for arbitrary projection methods. We evaluate three autoencoder (AE) architectures for creating parametric and invertible projections. Based on a given projection, we train AEs to learn a mapping into 2D space and an inverse mapping into the original space. We perform a quantitative and qualitative comparison on four datasets of varying dimensionality and pattern complexity using t-SNE. Our results indicate that AEs with a customized loss function can create smoother parametric and inverse projections than feed-forward neural networks while giving users control over the strength of the smoothing effect.

The VAST Challenge 2025 Mini-Challenge 3 focuses on identifying suspicious activities near Nemo Reef through the analysis of radio communications. We present RadioWave, a visual analytics application tailored to explore the knowledge graph inferred from these communications. RadioWave combines an aggregated node-link graph, a detailed non-aggregated graph, an interactive timeline, and a message view to support in-depth analysis. This combination allows users to identify groups, reveal pseudonyms, analyze temporal patterns and shifts, and inspect individual messages. Using RadioWave, we uncover two main opposing groups: one involved in suspicious rare earth mining and another aiming to protect Nemo Reef, and we show how their activities evolve.

One-dimensional orderings of spatial entities have been researched in many contexts, e.g. spatial indexing structures or visualizations for spatiotemporal trend analysis. While plenty of studies have been conducted to evaluate orderings of point-based data, polygonal shapes, despite their different topological properties, have received less attention. Existing measures to quantify errors in projections or orderings suffer from generic neighborhood definitions and over-simplification of distances when applied to polygonal data. In this work, we address these shortcomings by introducing measures that adapt to a varying neighborhood size depending on the number of contiguous neighbors and thus, address the limitations of existing measures for polygonal shapes. To guide experts in determining a suitable ordering, we propose a user-steerable visual analytics prototype capable of locally and globally inspecting ordering errors, investigating the impact of geographic obstacles, and comparing ordering strategies using our measures. We demonstrate the effectiveness of our approach through a use case and conducted an expert study with 8 data scientists as a qualitative evaluation of our approach. Our results show that users are capable of identifying ordering errors, comparing ordering strategies on a global and local scale, as well as assessing the impact of semantically relevant geographic obstacles.

The growing popularity of generative language models has amplified interest in interactive methods to guide model outputs. Prompt refinement is considered one of the most effective means to influence output among these methods. We identify several challenges associated with prompting large language models, categorized into data- and model-specific, linguistic, and socio-linguistic challenges. A comprehensive examination of model outputs, including runner-up candidates and their corresponding probabilities, is needed to address these issues. The beam search tree, the prevalent algorithm to sample model outputs, can inherently supply this information. Consequently, we introduce an interactive visual method for investigating the beam search tree, facilitating analysis of the decisions made by the model during generation. We quantitatively show the value of exposing the beam search tree and present five detailed analysis scenarios addressing the identified challenges. Our methodology validates existing results and offers additional insights.

The complexity of intervention studies to assess physical activity (PA) is increasing, resulting in vast amounts of data being recorded in laboratory settings. Recent studies extend datasets with measurements outside the lab using wearable devices, allowing for a bridge to be built between the lab and real-life applications. Such heterogeneous, multigranular datasets impose various challenges for data analysis and visualization, and require tailored approaches to support domain experts. Contrarily, it enables data-driven hypothesis generation, which is particularly valuable in interdisciplinary contexts where theory-driven approaches fall short due to a lack of well-established theoretical foundations. While lab conditions are extensively handled in existing visual interfaces, measurements in everyday life situations are often neglected, yet have an essential impact on the capturing of PA. To facilitate exploration of this data, we propose a visual analytics application consisting of multiple linked views comprising a BiPlot, Variable Distribution plots, and a dense-pixel visualization allowing experts to generate novel interdisciplinary hypotheses based on laboratory and everyday life measurements. We evaluate our application by conducting an expert study with an end user of the application, showcasing the application’s benefits to support experts in solving tasks regarding exploration, pattern identification, association, and comparison.

In Mini-Challenge 1 of the 2025 VAST Challenge, participants supported journalist Silas Reed in exploring the rise of the artist Sailor Shift and the global influence of Oceanus Folk music. We developed a visual analytics tool built on a Neo4j database, FastAPI backend, and React.js frontend with D3.js and Sigma.js for interactive graph and timeline visualizations. The tool enables exploration of artists, genres, and influences through linked visual components. It supports tracing collaborations, stylistic evolution, and the genre’s cultural diffusion over time. We found that Sailor Shift is rarely co-performing, but inspired many other artists with her songs and albums to create their own music. The genre of Oceanus Folk rose rather intermittent than gradually while influencing different genres like Desert Rock, Dream Pop and Indie Folk and artists like Min He and Xia Cui as well as taking inspirations from Alternative Rock, Desert Rock and Dream Pop. We visualized the careers of Filippo Pelli, Kimberly Snyders and Zacharie Martins and predicted Beatrice Albright, Daniel O’Connell, and Genevieve Bell to be the next rising stars based on our prediction tool.