Challenge: Prior work has shown that both humans and MLLMs are frequently deceived by misleading visualizations.
Approach: They propose a benchmark of 2,604 real-world visualizations annotated with 12 types of misleaders.
Outcome: The proposed framework can detect misleading visualizations and identify specific design rules they violate . the proposed framework is based on a synthetic dataset of 81,814 visualizations .

Similar Papers

Unmasking Deceptive Visuals: Benchmarking Multimodal Large Language Models on Misleading Chart Question Answering (2025.emnlp-main)

Copied to clipboard

Challenge: Misleading visualizations can distort perception and lead to incorrect conclusions.
Approach: They propose a large-scale multimodal dataset to evaluate MLLMs on misleading chart reasoning.
Outcome: The proposed framework evaluates MLLMs on misleading chart reasoning on a large-scale multimodal dataset spanning 21 misleader types and 10 chart types . it contains 3,026 curated examples spanning standard chart code, CSV data, multiple-choice questions, and labeled explanations, validated through iterative MLML checks and exhausted expert human review.
Protecting multimodal large language models against misleading visualizations (2026.acl-long)

Copied to clipboard

Challenge: MLLMs are robust to misleading visualizations, i.e., charts that distort the underlying data, leading readers to draw inaccurate conclusions.
Approach: They propose to use table-based QA and redrawing the visualization to improve QA performance on misleading visualizations.
Outcome: The proposed methods improve MLLM question-answering accuracy on misleading visualizations without compromising accuracy on non-misleading ones.
RADAR: A Reasoning-Guided Attribution Framework for Explainable Visual Data Analysis (2026.findings-eacl)

Copied to clipboard

Challenge: Multimodal Large Language Models (MLLMs) provide no visibility into which parts of visual data informed their conclusions.
Approach: They propose a semi-automatic approach to attribute reasoning process by highlighting regions in charts and graphs that justify model answers.
Outcome: The proposed method improves attribution accuracy by up to 15 percentage points compared to baseline methods and achieves high semantic similarity with ground truth responses.
ChartEdit: How Far Are MLLMs From Automating Chart Analysis? Evaluating MLLMs’ Capability via Chart Editing (2025.findings-acl)

Copied to clipboard

Challenge: Existing evaluations of multimodal large language models rely on limited case studies . however, they lack the ability to generate accurate edits according to the instructions .
Approach: They propose a benchmark for chart editing that includes 1,405 edit instructions applied to 233 real-world charts.
Outcome: The proposed benchmark includes 1,405 diverse editing instructions applied to 233 real-world charts.
ChartCheck: Explainable Fact-Checking over Real-World Chart Images (2024.findings-acl)

Copied to clipboard

Challenge: Data visualizations are often used to summarize and communicate key information, but they can also be misused to spread misinformation and promote agendas.
Approach: They propose a dataset for explainable fact-checking against real-world charts that uses vision-language and chart-to-table models to evaluate the validity of the dataset.
Outcome: The proposed model is based on vision-language and chart-to-table models and proposes a baseline to the community.
Not all Fake News is Written: A Dataset and Analysis of Misleading Video Headlines (2023.emnlp-main)

Copied to clipboard

Challenge: Social media platforms are used by half of U.S. adults for everyday news consumption.
Approach: They propose to analyze video headlines and whether annotators believe the headline is representative of the video’s contents.
Outcome: The proposed dataset analyzes video headlines and explains why annotators view a video as misleading.
Do LVLMs Understand Charts? Analyzing and Correcting Factual Errors in Chart Captioning (2024.findings-acl)

Copied to clipboard

Challenge: LVLMs are known for producing text that is factually inconsistent with visual input . factuality of generated captions for structured visuals has not been studied as much .
Approach: They propose a typology of factual errors in captions generated by large vision-language models . they propose CHOCOLATE, a visual entailment model that outperforms current models based on this analysis .
Outcome: The proposed model outperforms current models in evaluating caption factuality.
ChartCoder: Advancing Multimodal Large Language Model for Chart-to-Code Generation (2025.acl-long)

Copied to clipboard

Challenge: Existing open-source MLLMs fail to fully capture dense information embedded in charts . current models still face significant challenges in understanding and analyzing visual tasks such as captioning and question answering.
Approach: They propose a chart-to-code MLLM which leverages Code LLMs as the language backbone to enhance the executability of the generated code.
Outcome: The proposed model surpasses existing open-source models on chart-to-code benchmarks with only 7B parameters and provides lossless representations that contain all critical details.
Fact-Checking Meets Fauxtography: Verifying Claims About Images (D19-1)

Copied to clipboard

Challenge: Recent explosion of false claims in social media has led to manual fact-checking initiatives . however, existing methods are inadequate to deal with the growing number of false content claims.
Approach: They propose to model claims about images using a new dataset to examine the relationship between the image and the claim.
Outcome: The proposed method improves on the baseline and will enable future research on fact-checking claims about images.
Probing Multimodal Large Language Models on Cognitive Biases in Chinese Short-Video Misinformation (2026.findings-acl)

Copied to clipboard

Challenge: Short-video platforms have become major channels for misinformation, but their robustness against misinformation entangled with cognitive biases remains under-explored.
Approach: They propose a framework for evaluation of short-video platforms that use visual cues and social cue.
Outcome: The proposed framework evaluates MLLMs across five modality settings.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations