Search arXiv⌕ Search
Home/Archive/arXiv

arXiv · 2604.01181

True (VIS) Lies: Analyzing How Generative AI Recognizes Intentionality, Rhetoric, and Misleadingness in Visualization Lies

Graziano Blasilli·Marco Angelini
View original source ↗Read source document ↗Export metadata ↓

Abstract

This study investigates the ability of multimodal Large Language Models (LLMs) to identify and interpret misleading visualizations, and recognize these observations along with their underlying causes and potential intentionality. Our analysis leverages concepts from visualization rhetoric and a newly developed taxonomy of authorial intents as explanatory lenses. We formulated three research questions and addressed them experimentally using a dataset of 2,336 COVID-19-related tweets, half of which contain misleading visualizations, and supplemented it with real-world examples of perceptual, cognitive, and conceptual errors drawn from VisLies, the IEEE VIS community event dedicated to showcasing deceptive and misleading visualizations. To ensure broad coverage of the current LLM landscape, we evaluated 16 state-of-the-art models. Among them, 15 are open-weight models, spanning a wide range of model sizes, architectural families, and reasoning capabilities. The selection comprises small models, namely Nemotron-Nano-V2-VL (12B parameters), Mistral-Small-3.2 (24B), DeepSeek-VL2 (27B), Gemma3 (27B), and GTA1 (32B); medium-sized models, namely Qianfan-VL (70B), Molmo (72B), GLM-4.5V (108B), LLaVA-NeXT (110B), and Pixtral-Large (124B); and large models, namely Qwen3-VL (235B), InternVL3.5 (241B), Step3 (321B), Llama-4-Maverick (400B), and Kimi-K2.5 (1000B). In addition, we employed OpenAI GPT-5.4, a frontier proprietary model. To establish a human perspective on these tasks, we also conducted a user study with visualization experts to assess how people perceive rhetorical techniques and the authorial intentions behind the same misleading visualizations. This allows comparison between model and expert behavior, revealing similarities and differences that provide insights into where LLMs align with human judgment and where they diverge.

Explore related subjects

cs.HCcs.CLcs.CV

Keep this discovery

Explore connections, maps & timelines

BibTeXRIS

Graziano Blasilli, Marco Angelini. 2026-04-01. True (VIS) Lies: Analyzing How Generative AI Recognizes Intentionality, Rhetoric, and Misleadingness in Visualization Lies. https://arxiv.org/abs/2604.01181

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related papers

arXiv2026

Learning Password Best Practices Through In-Task Instruction

Users often make security- and privacy-relevant decisions without a clear understanding of the rules that govern safe behavior. We introduce pedagogical friction, a design approach that inserts brief, instructional interactions at the moment of action. We evaluate this approach in the context of password creation, a familiar task with clear quality criteria. We conducted a randomized study with 128 participants across four interface conditions that varied the depth and interactivity of guidance. We assessed three outcomes: (1) rule compliance in a subsequent password task without guidance, (2) accuracy on survey questions tied to password rules, and (3) behavior-knowledge alignment, which captures whether participants who correctly followed a rule also recognized it on the survey. Across the guided conditions, participants corrected most rule violations in the follow-up task and showed high behavior-knowledge alignment. Survey results suggested clearer advantages for some rule types, especially symbol related questions. These results position pedagogical friction as a lightweight intervention for security- and privacy-critical interfaces.

cs.HC↗
arXiv2026

Leveling the Playing Field: Temporal Video Segmentation for Individuals with ADHD in Computing Education

Individuals with Attention-Deficit/Hyperactivity Disorder (ADHD) often face significant barriers in computing education. In asynchronous learning environments, instructional videos can impose high extraneous cognitive load, often relying on assumptions about sustained attention and working memory that do not align with ADHD neurocognitive profiles. In this work, we evaluate a post-hoc video processing intervention that segments instructional content into single-instruction chunks followed by fixed-length pauses to reduce cognitive load. In a within-participants controlled study with 17 individuals with ADHD and 10 without, we find that the intervention has an equalizing effect. Although it improved performance for all participants, gains were larger for those with ADHD, reducing their errors and hesitations to levels comparable to those of participants without ADHD under the same intervention. These results align with the goals of Universal Design for Learning (UDL), by showing that cognitively-aligned, post-hoc instructional video modifications can reduce performance disparities across diverse neurocognitive profiles.

cs.HC↗
arXiv2026

Human Driver Temperament and the Safety Impact of a C-V2X Denial-of-Service Flooding Attack in Mixed-Autonomy Traffic

Cooperative and connected automated vehicles (CAVs) rely on Signal Phase and Timing (SPaT) messages to cross signalized intersections; a denial-of-service (DoS) flood that blocks SPaT forces CAVs into a fail-safe mode. Because human-driven vehicles share the intersection, the safety consequence depends not only on the attack and the CAV fail-safe policy, but on how the surrounding human drivers behave. We investigate this human-factors dimension with a coupled OMNeT++/INET (5G NR-V2X) and SUMO microsimulation of a signalized corridor, sweeping CAV market penetration (10-90%), four calibrated driver temperaments (cautious to aggressive) and two standards-based fail-safe policies, a minimal-risk maneuver (MRM) and an adaptive cruise control (ACC) keep-driving fallback, with each attack arm differenced against its policy-matched no-attack baseline. Temperament's effect on the attack is specific and modest rather than a blanket amplification. Aggressive surroundings worsen one metric, the hard-braking a keep-driving fail-safe forces on nearby drivers (p = 0.03), rising from near zero to +8 episodes/1000 veh-s. They appear to dampen rear-end conflicts, but only because the flood clears the queues aggressive drivers build, so the gain is in flow, not safety. On the attack's primary signatures, CAV red-light running and crossing conflicts, temperament has no detectable effect. It instead dominates baseline risk, producing a 13- to 18-fold cautious-to-aggressive gradient far larger than the attack itself, which acts through a channel already congested by CAV adoption. Human driver populations determine how dangerous the intersection is but do not systematically amplify this attack, so fail-safe design cannot assume a cautious test population bounds the risk.

cs.HC↗