Multimodal LLMs Can Reason about Aesthetics in Zero-Shot
s-emanuilov
Abstract
We present the first study on how Multimodal LLMs' (MLLMs) reasoning ability shall be elicited to evaluate the aesthetics of artworks. To facilitate this investigation, we construct MM-StyleBench, a novel high-quality dataset for benchmarking artistic stylization. We then develop a principled method for human preference modeling and perform a systematic correlation analysis between MLLMs' responses and human preference. Our experiments reveal an inherent hallucination issue of MLLMs in art evaluation, associated with response subjectivity. ArtCoT is proposed, demonstrating that art-specific task decomposition and the use of concrete language boost MLLMs' reasoning ability for aesthetics. Our findings offer valuable insights into MLLMs for art and can benefit a wide range of downstream applications, such as style transfer and artistic image generation. Code available at https://github.com/songrise/MLLM4Art.
Community
"Can MLLMs reason about the aesthetic quality of artistic images in a manner aligned with human preferences?"
A study from the Hong Kong Polytechnic University investigates how multimodal language models evaluate artistic aesthetics, introducing an ArtCoT prompting method that outperforms baseline approaches in aligning with human preferences.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- FiVL: A Framework for Improved Vision-Language Alignment (2024)
- Are They the Same? Exploring Visual Correspondence Shortcomings of Multimodal LLMs (2025)
- Task Preference Optimization: Improving Multimodal Large Language Models with Vision Task Alignment (2024)
- Socratic Questioning: Learn to Self-guide Multimodal Reasoning in the Wild (2025)
- ComprehendEdit: A Comprehensive Dataset and Evaluation Framework for Multimodal Knowledge Editing (2024)
- Maya: An Instruction Finetuned Multilingual Multimodal Model (2024)
- OCRBench v2: An Improved Benchmark for Evaluating Large Multimodal Models on Visual Text Localization and Reasoning (2024)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper