Elamur Research⌕
☰
home / papers / 2026 / Assessing the Persuasive Effect of AI-Generated Image Support of Arguments

Assessing the Persuasive Effect of AI-Generated Image Support of Arguments

Summary

The paper asks whether images made specifically for an argument can make that argument more persuasive. Earlier multimodal argument mining work used images that fit their texts only loosely, such as retrieved topic images, or images that mostly repeat written slogans, as in ImageArg tweets. To control the setup, the authors generated all material with generative AI. GPT-4 wrote pro and con arguments on five controversial topics, such as school uniforms and waste segregation, in two rhetorical styles: logos (reason-based) and pathos (emotion-based). Dall-E produced text-free candidate images for each side. Two user studies followed. The first selected the best image for each topic, stance and rhetorical strategy. The second compared how persuasive readers found arguments shown alone versus with the selected image. Images tended to raise perceived persuasiveness, more so for pathos-oriented arguments, but the effect varied by topic, possibly because some topics are harder to depict convincingly. The work gives an early basis for designing multimodal argumentation technologies.

Key points

  • Data set: 5 topics, each with 6 pro and 6 con arguments generated by GPT-4. Each side has 3 logos and 3 pathos arguments, following the rhetorical strategies of Wachsmuth et al. (2018).
  • Images: Dall-E generated 3 candidate images per topic and stance. The images contain no embedded text, so the visual effect can be studied on its own.
  • Study 1: 26 participants chose which candidate image best supported a logical view and which an emotional view of each topic and stance.
  • Study 2: The selected images were paired with 2 of the 3 arguments per topic, stance and strategy. Participants then compared text-only and text-plus-image persuasiveness. Images generally helped, with a stronger effect for pathos and clear differences between topics.
  • Resources: All arguments and images are public at github.com/discourse-lab/ImageArguments-LREC26. The authors used closed-source models for quality, which they acknowledge limits reproducibility.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/2b6xxx7vmyra

@inproceedings{quadras2026assessing,
  title = {{Assessing the Persuasive Effect of AI-Generated Image Support of Arguments}},
  author = {Quadras, Mackwyn and Stede, Manfred and Wachsmuth, Henning},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/2b6xxx7vmyra},
  url = {https://doi.org/10.63317/2b6xxx7vmyra}
}

References

31 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Khalid Al Khatib, Henning Wachsmuth, Johannes Kiesel, Matthias Hagen, and Benno Stein. 2016. A news editorial corpus for mining argumentation strategies. In Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, pages 3433–3443. find
  2. The COLING 2016 Organizing Committee. find
  3. Milad Alshomary, Roxanne El Baff, Timon Gurcke, and Henning Wachsmuth. 2022. The moral debater: A study on the computational generation of morally framed arguments. In Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 8782–8797, Du… find
  4. Pranav Anand, Joseph King, Jordan Boyd-Graber, Earl Wagner, Craig Martell, Doug Oard, and Philip Resnik. 2011. Believe me: we can do this! annotating persuasive acts in blog text. In Proceedings of the 10th AAAI Conference on Computational Models of Natural Argument, AAAIWS’11-10, page 11?15. AAAI P… find
  5. Aristotle. ca. 350 B.C.E./ translated 2007. On Rhetoric: A Theory of Civic Discourse. Oxford University Press, Oxford, UK. Translated by George A. Kennedy. find
  6. Miriam Louise Carnot, Lorenz Heinemann, Jan Braker, Tobias Schreieder, Johannes Kiesel, Maik Fröbe, Martin Potthast, and Benno Stein. 2023. On stance detection in image retrieval for argumentation. In 46th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR… find
  7. Jiwan Chung, Sungjae Lee, Minseo Kim, Seungju Han, Ashkan Yousefpour, Jack Hessel, and Youngjae Yu. 2024. Selective vision is the challenge for visual reasoning: A benchmark for visual argument understanding. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing,… find
  8. Rory Duthie, Katarzyna Budynska, and Chris Reed. 2016. Mining ethos in political debate. In Proceedings of the Sixth International Conference on Computational Models of Argument, pages 299–310. find
  9. Roxanne El Baff, Henning Wachsmuth, Khalid Al Khatib, Manfred Stede, and Benno Stein. 2019. Computational argumentation synthesis as a language modeling task. In Proceedings of the 12th International Conference on Natural Language Generation, pages 54–64, Tokyo, Japan. Association for Computational… find
  10. Roxanne El Baff, Henning Wachsmuth, Khalid Al Khatib, and Benno Stein. 2018. Challenge or empower: Revisiting argumentation quality in a news editorial corpus. In Proceedings of the 22nd Conference on Computational Natural Language Learning, pages 454–464. Association for Computational Linguistics. find
  11. Roxanne El Baff, Henning Wachsmuth, Khalid Al Khatib, and Benno Stein. 2020. Analyzing the persuasive effect of style in news editorial argumentation. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 3154–3160, Online. Association for Computational Li… find
  12. Johannes Kiesel, Nico Reichenbach, Benno Stein, and Martin Potthast. 2021. Image retrieval for arguments using stance-aware query expansion. In Proceedings of the 8th Workshop on Argument Mining, pages 36–45, Punta Cana, Dominican Republic. Association for Computational Linguistics. find
  13. John Lawrence and Chris Reed. 2019. Argument mining: A survey. Computational Linguistics, 45(4):765–818. find our card
  14. Zhexiong Liu, Mohamed Elaraby, Yang Zhong, and Diane Litman. 2023. Overview of ImageArg-2023: The first shared task in multimodal argument mining. In Proceedings of the 10th Workshop on Argument Mining, pages 120–132, Singapore. Association for Computational Linguistics. find
  15. Zhexiong Liu, Meiqi Guo, Yue Dai, and Diane Litman. 2022. ImageArg: A multi-modal tweet dataset for image persuasiveness mining. In Proceedings of the 9th Workshop on Argument Mining, pages 1–18, Online and in Gyeongju, Republic of Korea. International Conference on Computational Linguistics. find
  16. Stephanie Lukin, Pranav Anand, Marilyn Walker, and Steve Whittaker. 2017. Argument strength is in the eye of the beholder: Audience effects in persuasion. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 742–… find
  17. Eleonora Mancini, Federico Ruggeri, Stefano Colamonaco, Andrea Zecca, Samuele Marro, and Paolo Torroni. 2024. MAMKit: A comprehensive multimodal argument mining toolkit. In Proceedings of the 11th Workshop on Argument Mining (ArgMining 2024), pages 69–82, Bangkok, Thailand. Association for Computati… find
  18. Lily Ng, Anne Lauscher, Joel Tetreault, and Courtney Napoles. 2020. Creating a domain-diverse corpus for theory-based argument quality assessment. In Proceedings of the 7th Workshop on Argument Mining, pages 117–126, Online. Association for Computational Linguistics. find
  19. Daniel J. O’Keefe. 2002. Persuasion: Theory and research, 2nd edition. Sage Publications. find
  20. OpenAI. 2023. GPT-4 technical report. find our card
  21. Andreas Peldszus and Manfred Stede. 2016. An annotated corpus of argumentative microtexts. In Argumentation and Reasoned Action: Proceedings of the 1st European Conference on Argumentation, Lisbon 2015 / Vol. 2, pages 801–816. College Publications, London. find
  22. Aditya Ramesh, Mikhail Pavlov, Gabriel Goh, Scott Gray, Chelsea Voss, Alec Radford, Mark Chen, and Ilya Sutskever. 2021. Zero-shot text-to-image generation. find our card
  23. Gabriella Skitalinskaya, Jonas Klaff, and Henning Wachsmuth. 2021. Learning from revisions: Quality assessment of claims in argumentation at scale. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 1718–1729, Online. As… find
  24. Manfred Stede, Jodi Schneider, and Henning Wachsmuth. 2026. Argument Mining (Second Edition). Synthesis Lectures on Human Language Technologies. Springer, Cham. find
  25. Assaf Toledo, Shai Gretz, Edo Cohen-Karlik, Roni Friedman, Elad Venezian, Dan Lahav, Michal Jacovi, Ranit Aharonov, and Noam Slonim. 2019. Automatic argument quality assessment - New datasets and methods. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and t… find
  26. Stephen Toulmin. 1958. The Uses of Argument. Cambridge University Press, Cambridge. find
  27. Frans H. van Eemeren and Rob Grootendorst. 2004. A Systematic Theory of Argumentation: The Pragma-Dialectical Approach. Cambridge University Press, Cambridge, UK. find
  28. Henning Wachsmuth, Nona Naderi, Yufang Hou, Yonatan Bilu, Vinodkumar Prabhakaran, Tim Alberdingk Thijm, Graeme Hirst, and Benno Stein. 2017. Computational argumentation quality assessment in natural language. In Proceedings of the 15th Conference of the European Chapter of the Association for Comput… find our card
  29. Henning Wachsmuth, Manfred Stede, Roxanne El Baff, Khalid Al Khatib, Maria Skeppstedt, and Benno Stein. 2018. Argumentation synthesis following rhetorical strategies. In Proceedings of the 27th International Conference on Computational Linguistics, pages 3753–3765. Association for Computational Ling… find
  30. D. Walton, C. Reed, and F. Macagno. 2008. Argumentation schemes. Cambridge University Press, Cambridge. find
  31. Xuewei Wang, Weiyan Shi, Richard Kim, Yoojung Oh, Sijia Yang, Jingwen Zhang, and Zhou Yu. 2019. Persuasion for good: Towards a personalized persuasive dialogue system for social good. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5635–5649, Florenc… find our card

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/quadras-2026-assessing-persuasive-effect-ai-generated/