Understanding How MLLMs Describe Artworks Using Token Activation Maps
- 1 October 2026
- book chapter
- Published by Springer Nature
- p. 124-138
Abstract
No abstract availableKeywords
All Related Versions
This publication has 26 references indexed in Scilit:
- Have Large Vision-Language Models Mastered Art History?2026
- Does CLIP Perceive Art the Same Way We Do?2025
- GraphCLIP: Image-graph contrastive learning for multimodal artwork classificationKnowledge-Based Systems, 2024
- Rotary Position Embedding for Vision Transformer2024
- Multimodal Large Language Models: A Survey2023
- From attribution maps to human-understandable explanations through Concept Relevance PropagationNature Machine Intelligence, 2023
- A Dataset and Baselines for Visual Question Answering on Art2020
- A Deep Learning Perspective on Beauty, Sentiment, and Remembrance of ArtIEEE Access, 2019
- How to Read Paintings: Semantic Art Understanding with Multi-modal Retrieval2019
- Learning Deep Features for Discriminative Localization2016