arXiv subjects

Lonni Besançon

Publications and source records attributed to Lonni Besançon.

18 recordsLinked to original sources

arXiv2026

More Than 63% of IEEE VIS Research Liable to be Retracted?! Ethics Approval Statements Protect Participants (and Researchers!)

We analyzed the ethics reporting in 255 IEEE VIS papers from 2024 and 2025, as published in TVCG. This analysis arose from our experience as readers and reviewers of IEEE VIS papers that such reporting is frequently incomplete or missing, as well as from investigations in which we ourselves had to answer challenges regarding ethics approval in our own work. Visualization research naturally often involves human participants, yet ethics approval and informed-consent procedures are not always explicitly reported. In our corpus, 189 papers (74.1%) reported on work involving human participants. Only 6 of them (3.2%) reported to have obtained ethics approval, an approval identifier, and having received informed consent from the participants, while 26 (13.8%) reported at least ethics approval and informed consent. These omissions do not imply that the empirical work was unethical or lacked approval. They rather show that current reporting practices make ethical approval difficult to assess. Beyond our alarming - yet true - paper title, we wish to raise awareness on how authors themselves may eventually be at risk for not properly reporting ethics. We argue that VIS should adopt clearer and more standardized ethics-reporting practices to protect participants, authors, reviewers, and editors.

cs.HC↗

arXiv2026

State of the Art of LLM-Enabled Interaction with Visualization

We report on a systematic, PRISMA-guided survey of research at the intersection of LLMs and visualization, with a particular focus on visio-verbal interaction -- where verbal and visual modalities converge to support data sense-making. The emergence of Large Language Models (LLMs) has introduced new paradigms for interacting with data visualizations through natural language, leading to intuitive, multimodal, and accessible interfaces. We analyze 48 papers across six dimensions: application domain, visualization task, visualization representation, interaction modality, LLM integration, and system evaluation. Our classification framework maps LLM roles across the visualization pipeline, from data querying and transformation to visualization generation, explanation, and navigation. We highlight emerging design patterns, identify gaps in accessibility and visualization reading, and discuss the limitations of current LLMs in spatial reasoning and contextual grounding. We further reflect on evaluations of combined LLM-visualization systems, highlighting how current research projects tackle this challenge and discuss current gaps in conducting meaningful evaluations of such systems. With our survey we aim to guide future research and system design in LLM-enhanced visualization, supporting broad audiences and intelligent, conversational interfaces.

cs.HC↗

arXiv2026

Piloting Planetarium Visualizations with LLMs during Live Events in Science Centers

We designed and evaluated an AI pilot in a planetarium visualization software, OpenSpace, for public shows in science centers. The piloting role is usually given to a human working in close collaboration with the guide on stage. We recruited 7 professional guides with extensive experience in giving shows to the public to study the impact of the AI-piloting on the overall experience. The AI-pilot is a conversational AI-agent listening to the guide and interpreting the verbal statements as commands to execute camera motions, change simulation time, or toggle visual assets. Our results show that, while AI pilots lack several critical skills for live shows, they could become useful as co-pilots to reduce workload of human pilots and allow multitasking. We propose research directions toward implementing visualization pilots and co-pilots in live settings.

cs.HC↗

arXiv2026

Opportunities of Touch-Enabled Spherical Displays to support Climate Conversations

We explore how touch-sensitive spherical displays can support climate conversations in museums and science centers. These displays enable intuitive and embodied interaction with complex climate data, and support collective exploration. However, current interaction capabilities of spherical displays are limited. Therefore, this exploratory study aims to identify potential opportunities to develop meaningful interactions and technical solutions. Through two workshops, key opportunities were identified to improve visitors' understanding and navigation of climate data, along with recommendations for technical implementation. Our results provide guidelines and aspects to consider for future research and development in this area.

cs.HC↗

arXiv2026

MarioChart: Autonomous Tangibles as Active Proxy Interfaces for Embodied Casual Data Exploration

We introduce the notion of an Active Proxy interface, i.e. tangible models as proxies for physical data referents, supporting interactive exploration of data through active manipulation. We realise an active proxy data visualisation system, "MarioChart", using robot carts relocating themselves on a tabletop, e.g., to align with their data referents in a map or other visual layout. We consider a casual-data exploration scenario involving a multivariate campus sustainability dataset, using scale models as proxies for their physical building data referents. Our empirical study (n=12) compares active proxy use with conventional tablet interaction, finding that our active proxy system enhances short-term spatial memory of data and enables faster completion of certain data analytic tasks. It shows no significant differences compared to traditional touch-screens in long-term memory, physical fatigue, mental workload, or user engagement. Our study offers an initial baseline for active proxy techniques and advances understanding of tangible interfaces in situated data visualisation.

cs.HC↗

arXiv2026

Criminator: An Easy-to-Use XR "Crime Animator" for Rapid Reconstruction and Analysis of Dynamic Crime Scenes

Law enforcement authorities are increasingly interested in 3D modelling for virtual crime scene reconstruction, enabling offline analysis without the cost and contamination risk of on-site investigation. Past work has demonstrated spatial relationships through static modelling but validating the sequence of events in dynamic scenarios is crucial for solving a case. Yet, animation tools are not well suited to crime scene reconstruction, and complex for non-experts in 3D modelling/animation. Through a co-design process with criminology experts, we designed "Criminator"-a methodological framework and XR tool that simplifies animation authoring. We evaluated this tool with participants trained in criminology (n=6) and untrained individuals (n=12). Both groups were able to successfully complete the character animation tasks and provided high usability ratings for observation tasks. Criminator has potential for hypothesis testing, demonstration, sense-making, and training. Challenges remain in how such a tool fits into the entire judicial process, with questions about including animations as evidence.

cs.HC↗

arXiv2026

Children's Expectations, Engagement, and Evaluation of an LLM-enabled Spherical Visualization Platform in the Classroom

We present our first stage results from deploying an LLM-augmented visualization software in a classroom setting to engage primary school children with earth-related datasets. Motivated by the growing interest in conversational AI as a means to support inquiry-based learning, we investigate children's expectations, engagement, and evaluation of a spoken LLM interface with a shared, immersive visualization system in a formal educational context. Our system integrates a speech-capable large language model with an interactive spherical display. It enables children to ask natural-language questions and receive coordinated verbal explanations and visual responses through the LLM-augmented visualization updating in real time based on spoken queries. We report on a classroom study with Swedish children aged 9-10, combining structured observation and small-group discussions to capture expectations prior to interaction, interaction patterns during facilitated sessions, and children's reflections on their encounter afterward. Our results provide empirical insights into children's initial encounters with an LLM-enabled visualization platform within a classroom setting and their expectations, interactions, and evaluations of the system. These findings inform the technology's potential for educational use and highlight important directions for future research.

cs.HC↗

arXiv2025

Lessons Learnt from Expert-Centred Studies Exploring Opportunities and Challenges for Immersive Forensic Investigation

Research studies involving human participants present challenges, including strict ethical considerations, participant recruitment, costs, and many human factors. While human-computer interaction researchers are familiar with these challenges and current solutions, expert-centred studies can be even more challenging in ways that researchers may not anticipate. This issue is particularly important as research grants are increasingly based on practical and real-world problems, which necessitate close collaboration with experts. In this paper, we reflect on and discuss the challenges, solutions, and specific requirements that arose during our expert-centred studies conducted over three years of a PhD study exploring immersive forensic investigation.

cs.HC↗

arXiv2025

From "Arbitrary Timberland" To "Skyline Charts": Is Visualization At Risk From The Pollution of Scientific Literature?

In this essay, I argue that, while visualization research does not seem to be directly at risk of being corrupted by the current massive wave of polluted research, certain visualization concepts are being used in fraudulent fashions and fields close to ours are being targeted. Worse, the society publishing our work is overwhelmed by thousands of questionable papers that are being, unfortunately, published. As a community, and if we want our research to remain as good as it currently is, I argue that we should all get involved with our variety of skills to help identify and correct the current scientific record. I thus aim to present a few questionable practices that are worth knowing about when reviewing for fields using visualization research, and hopefully will never be useful when reviewing for our main venues. I also argue that our skill set could become particularly relevant in the future and invite scholars of the fields to try to get involved.

cs.HC↗

arXiv2025

Augmenting a Large Language Model with a Combination of Text and Visual Data for Conversational Visualization of Global Geospatial Data

We present a method for augmenting a Large Language Model (LLM) with a combination of text and visual data to enable accurate question answering in visualization of scientific data, making conversational visualization possible. LLMs struggle with tasks like visual data interaction, as they lack contextual visual information. We address this problem by merging a text description of a visualization and dataset with snapshots of the visualization. We extract their essential features into a structured text file, highly compact, yet descriptive enough to appropriately augment the LLM with contextual information, without any fine-tuning. This approach can be applied to any visualization that is already finally rendered, as long as it is associated with some textual description.

cs.HC↗

arXiv2025

Detection of metadata manipulations: Finding sneaked references in the scholarly literature

We report evidence of a new set of sneaked references discovered in the scientific literature. Sneaked references are references registered in the metadata of publications without being listed in reference section or in the full text of the actual publications where they ought to be found. We document here 80,205 references sneaked in metadata of the International Journal of Innovative Science and Research Technology (IJISRT). These sneaked references are registered with Crossref and all cite -- thus benefit -- this same journal. Using this dataset, we evaluate three different methods to automatically identify sneaked references. These methods compare reference lists registered with Crossref against the full text or the reference lists extracted from PDF files. In addition, we report attempts to scale the search for sneaked references to the scholarly literature.

cs.DL↗

arXiv2023

LSDvis: Hallucinatory Data Visualisations in Real World Environments

We propose the concept of "LSDvis": the (highly exaggerated) visual blending of situated visualisations and the real-world environment to produce data representations that resemble hallucinations. Such hallucinatory visualisations incorporate elements of the physical environment, twisting and morphing their appearance such that they become part of the visualisation itself. We demonstrate LSDvis in a ``proof of proof of concept'', where we use Stable Diffusion to modify images of real environments with abstract data visualisations as input. We conclude by discussing considerations of LSDvis. We hope that our work promotes visualisation designs which deprioritise saliency in favour of quirkiness and ambience.

cs.HC↗

arXiv2023

Sneaked references: Cooked reference metadata inflate citation counts

We report evidence of an undocumented method to manipulate citation counts involving 'sneaked' references. Sneaked references are registered as metadata for scientific articles in which they do not appear. This manipulation exploits trusted relationships between various actors: publishers, the Crossref metadata registration agency, digital libraries, and bibliometric platforms. By collecting metadata from various sources, we show that extra undue references are actually sneaked in at Digital Object Identifier (DOI) registration time, resulting in artificially inflated citation counts. As a case study, focusing on three journals from a given publisher, we identified at least 9% sneaked references (5,978/65,836) mainly benefiting two authors. Despite not existing in the articles, these sneaked references exist in metadata registries and inappropriately propagate to bibliometric dashboards. Furthermore, we discovered 'lost' references: the studied bibliometric platform failed to index at least 56% (36,939/65,836) of the references listed in the HTML version of the publications. The extent of the sneaked and lost references in the global literature remains unknown and requires further investigations. Bibliometric platforms producing citation counts should identify, quantify, and correct these flaws to provide accurate data to their patrons and prevent further citation gaming.

cs.DL↗

arXiv2021

Can visualization alleviate dichotomous thinking? Effects of visual representations on the cliff effect

Common reporting styles for statistical results in scientific articles, such as p-values and confidence intervals (CI), have been reported to be prone to dichotomous interpretations, especially with respect to the null hypothesis significance testing framework. For example when the p-value is small enough or the CIs of the mean effects of a studied drug and a placebo are not overlapping, scientists tend to claim significant differences while often disregarding the magnitudes and absolute differences in the effect sizes. This type of reasoning has been shown to be potentially harmful to science. Techniques relying on the visual estimation of the strength of evidence have been recommended to reduce such dichotomous interpretations but their effectiveness has also been challenged. We ran two experiments on researchers with expertise in statistical analysis to compare several alternative representations of confidence intervals and used Bayesian multilevel models to estimate the effects of the representation styles on differences in researchers' subjective confidence in the results. We also asked the respondents' opinions and preferences in representation styles. Our results suggest that adding visual information to classic CI representation can decrease the tendency towards dichotomous interpretations - measured as the `cliff effect': the sudden drop in confidence around p-value 0.05 - compared with classic CI visualization and textual representation of the CI with p-values. All data and analyses are publicly available at https://github.com/helske/statvis.

stat.OT↗

arXiv2020

An Evaluation of Visualization Methods for Population Statistics Based on Choropleth Maps

We evaluate several augmentations to the choropleth map to convey additional information, including glyphs, 3D, cartograms, juxtaposed maps, and shading methods. While choropleth maps are a common method used to represent societal data, with multivariate data they can impede as much as improve understanding. In particular large, low population density regions often dominate the map and can mislead the viewer as to the message conveyed. Our results highlight the potential of 3D choropleth maps as well as the low accuracy of choropleth map tasks with multivariate data. We also introduce and evaluate popcharts, four techniques designed to show the density of population at a very fine scale on top of choropleth maps. All the data, results, and scripts are available from https://osf.io/8rxwg/.

cs.HC↗

arXiv2016

Preference Between Allocentric and Egocentric 3D Manipulation in a Locally Coupled Configuration

We study user preference between allocentric and egocentric 3D manipulation on mobile devices, in a configuration where the motion of the device is applied to an object displayed on the device itself. We first evaluate this preference for translations and for rotations alone, then for full 6-DOF manipulation. We also investigate the role of contextual cues by performing this experiment in different 3D scenes. Finally, we look at the specific influence of each manipulation axis. Our results provide guidelines to help interface designers select an appropriate default mapping in this locally coupled configuration.

cs.HC↗

arXiv2016

Usability Comparison of Mouse, Touch and Tangible Inputs for 3D Data Manipulation

We evaluate the performance and usability of mouse-based, touch-based, and tangible interaction for manipulating objects in a 3D virtual environment. This comparison is a step toward a better understanding of the limitations and benefits of these existing interaction techniques, with the ultimate goal of facilitating the integration of different 3D data exploration environments into a single interaction continuum. For this purpose we analyze participants' performance in 3D manipulation using a docking task. We measured completion times, docking precision, as well as subjective criteria such as fatigue, workload, and preference. Our results show that the three input modalities provide similar levels of precision but require different interaction times. We also discuss our qualitative observations as well as people's preferences and put our findings into context of the practical application domain of 3D data analysis environments.

cs.HC↗

arXiv2016

A Tangible Volume for Portable 3D Interaction

We present a new approach to achieve tangible object manipulation with a single, fully portable and self-contained device. Our solution is based on the concept of a "tangible volume". We turn a tangible object into a handheld fish-tank display. The tangible volume represents a volume of space that can be freely manipulated within a virtual scene. This volume can be positioned onto virtual objects to directly grasp them, and to manipulate them in 3D space. We investigate this concept through two user studies. The first study evaluates the intuitiveness of using a tangible volume for grasping and manipulating virtual objects. The second study evaluates the effects of the limited field of view on spatial awareness. Finally, we present a generalization of this concept to other forms of interaction through the surface of the volume.

cs.HC↗