Sense it with your eyes: Sensation Generation and Understanding for Advertisements

2026-07-28

AI Summary

This study addresses the lack of systematic investigation and evaluation methods regarding how visual cues in advertising effectively evoke human sensory experiences. To bridge this gap, the authors introduce the first sensory advertising dataset, propose a novel sensory classification task (SenseClass), and develop an automated evaluation metric, SenseScore. Furthermore, they design SAGA, a multi-agent generative framework that integrates large language models with multimodal foundation models to enable understanding, evaluation, and generation of sensory-rich advertisements. Experimental results demonstrate that SenseScore aligns closely with human judgments, and SAGA significantly enhances the sensory evocativeness, image-text alignment, and persuasive power of generated ads, thereby establishing a foundational approach for research on sensory-driven visual persuasion.

Abstract

Sensory advertising evokes human senses through visual cues, enabling audiences to mentally simulate experiences and increasing persuasive impact. Despite the recent increase in using AI in generating and understanding creative and persuasive content, how advertisements visually evoke sensations remains largely unexplored. In this work, we introduce the first study of understanding, evaluating, and generating sensory ads. We introduce the Sensory Ad dataset, and define sensation classification tasks (SenseClass) to benchmark LLMs and MLLMs. We further propose SenseScore, an automated evaluation metric for sensation evocation achieving strong agreement with human judgments. Finally, we introduce the Sensory Ad Generation (SenseGen) task and propose SAGA, a multi-agent framework that improves message image alignment, sensory evocation, and persuasion. Our work establishes a foundation for sensory-aware visual persuasion.

Problem

• sensory advertising
• sensation evocation
• visual persuasion
• ad generation
• multimodal understanding

Innovation

• Sensory Advertising
• Multimodal Large Language Models
• Automated Evaluation Metric
• Multi-agent Framework
• Visual Persuasion