Visual Interestingness Decoded: How GPT-4o Mirrors Human Interests
Not on the map. Why?
Abstract (arXiv)
Our daily life is highly influenced by what we consume and see. Attracting and holding one's attention -- the definition of (visual) interestingness -- is essential. The rise of Large Multimodal Models (LMMs) trained on large-scale visual and textual data has demonstrated impressive capabilities. We explore these models' potential to understand to what extent the concepts of visual interestingness are captured and examine the alignment between human assessments and GPT-4o's, a leading LMM, predictions through comparative analysis. Our studies reveal partial alignment between humans and GPT-4o. It already captures the concept as best compared to state-of-the-art methods. Hence, this allows for the effective labeling of image pairs according to their (commonly) interestingness, which are used as training data to distill the knowledge into a learning-to-rank model. The insights pave the way for a deeper understanding of human interest.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 21 of its 41 references that are in the map. First posted on arXiv 15 Oct 2025.
Builds on
21 of 41 references.
References · 21
- 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
- 2018UMAP: Uniform Manifold Approximation and Projection for Dimension Reductioncited by 2,841
- 2022Locating and Editing Factual Associations in GPTcited by 2,609
- 2025Janus-Pro: Unified Multimodal Understanding and Generation with Data and Model Scalingcited by 787
- 2024DeepSeek-VL2: Mixture-of-Experts Vision-Language Models for Advanced Multimodal Understandingcited by 559
- 2024A Survey on Knowledge Distillation of Large Language Modelscited by 276
- 2022What does a platypus look like? Generating customized prompts for zero-shot image classificationcited by 263
- 2022Towards Understanding Grokking: An Effective Theory of Representation Learningcited by 229
- 2024GPT-4V(ision) is a Human-Aligned Evaluator for Text-to-3D Generationcited by 151
- 2024DreamBench++: A Human-Aligned Benchmark for Personalized Image Generationcited by 131
9 more
- 2024Towards Open-ended Visual Quality Comparisoncited by 85
- 2023VILA: Learning Image Aesthetics from User Comments with Vision-Language Pretrainingcited by 63
- 2023Bridging the Human-AI Knowledge Gap: Concept Discovery and Transfer in AlphaZerocited by 41
- 2024Can GPTs Evaluate Graphic Design Based on Design Principles?cited by 14
- 2021Embracing New Techniques in Deep Learning for Estimating Image Memorabilitycited by 10
- 2015Understanding Image Viralitycited by 7
- 2024Commonly Interesting Imagescited by 3
- 2024An Empirical Analysis of GPT-4V's Performance on Fashion Aesthetic Evaluationcited by 3
- 2024AID-AppEAL: Automatic Image Dataset and Algorithm for Content Appeal Enhancement and Assessment Labelingcited by 1
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2026Would you still call this Dax? Novel Visual References in VLMs and Humans
- 2023The Dawn of LMMs: Preliminary Explorations with GPT-4V(ision)
- 2026Human-Like Attention? A Psychophysical Comparison of Visual Search in Humans and MLLMs
- 2025I Spy With My Model's Eye: Visual Search as a Behavioural Test for MLLMs
- 2025From Pixels to Feelings: Aligning MLLMs with Human Cognitive Perception of Images
- 2025Vision Language Models as Values Detectors
- 2023Exploring Recommendation Capabilities of GPT-4V(ision): A Preliminary Case Study
- 2026Similar Choices, Different Attention: Cross-Modal Associations in Humans and Vision-Language Models