New framework for sensory ad generation and understanding introduced

Researchers have introduced a new framework for understanding and generating sensory advertisements, aiming to enhance their persuasive impact. This work includes the creation of the Sensory Ad dataset and the development of SenseClass, a task designed to benchmark large language models (LLMs) and multimodal LLMs (MLLMs) on their ability to classify sensory cues in ads. Additionally, a novel evaluation metric called SenseScore has been proposed, which correlates well with human judgment, and a multi-agent framework named SAGA has been developed for generating sensory advertisements that improve message-image alignment and persuasion. AI

IMPACT Establishes a foundation for sensory-aware visual persuasion, potentially improving AI-driven advertising.

RANK_REASON Research paper introducing a new dataset, tasks, and framework for sensory advertisements. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• Aysan Aghazadeh
• Hugging Face
• LLMs
• MLLMs
• SAGA
• SenseClass
• Sense it with your eyes: Sensation Generation and Understanding for Advertisements
• SenseScore
• Sensory Ad dataset

• paper
• product

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →