Research map / Probabilistic and scientific modeling / Causal discovery
Understanding Perceptual and Conceptual Fluency at a Large Scale
Shengli Hu, Ali Borji
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| fine-tune last three layers instead of transfer learning only | ResNet50 with ImageNet init · Logo design rank correlation | rank correlation: 0.50 -> 0.58 | Improved |
| fine-tune last three layers instead of transfer learning only | InceptionV3 with random init · Logo design rank correlation | rank correlation: 0.50 -> 0.54 | Improved |
| use ImageNet instead of random weight initialization (ablation) | ResNet50 fine-tuned · Logo design rank correlation | rank correlation: 0.58 -> 0.58 | No clear effect |
| use ImageNet instead of random weight initialization (ablation) | AlexNet transfer · Logo design rank correlation | rank correlation: 0.37 -> 0.41 | Improved |
| use ImageNet instead of random weight initialization (ablation) | VGG16 transfer · Logo design rank correlation | rank correlation: 0.40 -> 0.38 | Hurt |
| replace AlexNet with ResNet50 backbone | fine-tuned, ImageNet init · Logo design rank correlation | rank correlation: 0.39 -> 0.58 | Improved |
| increase perceptual complexity | linear regression on memorability · Logo memorability regression | coefficient: 0.290 | Improved |
| increase perceptual complexity | linear regression on likability · Logo likability regression | coefficient: -0.027 | Hurt |
| increase conceptual entropy (more ambiguous) | linear regression on memorability · Logo memorability regression | coefficient: -0.018 | Hurt |
| increase conceptual KL divergence | linear regression on memorability · Logo memorability regression | coefficient: 0.006 | Improved |
| add Perceptual Complexity squared term | regression on memorability · Logo memorability regression | coefficient: -0.074 | Hurt |
| interaction of PPI and conceptual entropy | regression on likability · Logo likability regression | coefficient: 0.006 | Improved |
Closest papers by meaning
- Can Vision–Language Models Assess Graphic Design Aesthetics? A Benchmark, Evaluation, and Dataset Perspective. (ICLR 2026 · similarity 0.84)
- Cross-modal Associations in Vision and Language Models: Revisiting the Bouba-Kiki Effect (NeurIPS 2025 · similarity 0.83)
- Synthesizing Images on Perceptual Boundaries of ANNs for Uncovering and Manipulating Human Perceptual Variability (ICML 2025 · similarity 0.83)
- Within-Model vs Between-Prompt Variability in Large Language Models for Creative Tasks (NeurIPS 2026 · similarity 0.83)
- Latent Representation Matters: Human-like Sketches in One-shot Drawing Tasks (NeurIPS 2024 · similarity 0.82)
- Bridging Cognitive Gap: Hierarchical Description Learning for Artistic Image Aesthetics Assessment (AAAI 2026 · similarity 0.81)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
