aiqbench analysis
This research introduces UkrMeme, a specialized dataset and benchmark for detecting persuasive techniques in Ukrainian wartime memes using vision-language models. The study evaluates zero-shot performance across multimodal inputs and compares interpretability methods like Layer-wise Relevance Propagation to understand how AI models attribute evidence in politically charged visual-textual contexts. It advances the field of explainable AI by addressing a critical niche where cultural nuance, language, and visual persuasion intersect during active conflict.
UkrMeme: Multimodal Explainable Persuasion Detection in Ukrainian Memes
arXiv:2610.13783v1 Announce Type: cross Abstract: Memes play a central role in modern online communication, especially in contexts of political conflict where visual and textual cues combine to convey persuasive messages. We study persuasive technique detection in Ukrainian memes using vision-language models (VLMs). We introduce UkrMeme, a dataset of 200 organically produced Ukrainian wartime memes annotated for persuasion techniques, and PTM-UA, a Ukrainian translation of a subset of the Propaganda 950 benchmark that preserves original visual content while translating meme text. We evaluate VLMs in a zero-shot prompting setup across image, text, and multimodal inputs using hierarchical F1 metrics. Finally, we compare Layer-wise Relevance Propagation (LRP) and model-generated token attributions and analyze the semantic categories of evidence emphasized by each explanation method.
Abstract:Memes play a central role in modern online communication, especially in contexts of political conflict where visual and textual cues combine to convey persuasive messages. We study persuasive technique detection in Ukrainian memes using vision-language models (VLMs). We introduce UkrMeme, a dataset of 200 organically produced Ukrainian wartime memes annotated for persuasion techniques, and PTM-UA, a Ukrainian translation of a subset of the Propaganda 950 benchmark that preserves original visual content while translating meme text. We evaluate VLMs in a zero-shot prompting setup across image, text, and multimodal inputs using hierarchical F1 metrics. Finally, we compare Layer-wise Relevance Propagation (LRP) and model-generated token attributions and analyze the semantic categories of evidence emphasized by each explanation method.
Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computer Vision and Pattern Recognition (cs.CV)
arXiv:2610.13783 [cs.AI]
(or arXiv:2610.13783v1 [cs.AI] for this version)
https://doi.org/10.48550/arXiv.2610.13783 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 220 words
Related coverage
Beyond a Joke: Multi-Angle Reasoning for Detecting and Explaining Harmful Humor in Memes
arXiv · Jul 20, 2026
Learn Before You Judge: Progressive Knowledge-to-Decision Alignment for Explainable Hateful Meme Detection
arXiv · Sep 18, 2026
[AINews] High Taste Testers Ep. 1
Latent Space · Oct 17, 2026
[AINews] not much happened today
Latent Space · Sep 10, 2026