Research map / Vision-language and multimodal / Human and embodied modeling
Understanding Aesthetics with Language: A Photo Critique Dataset for Aesthetic Assessment
Daniel Vera Nieto, Luigi Celona, Clara Fernandez Labrador
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 |
|---|---|---|---|
| Use ViT-L/16 pre-trained on ImageNet-21k instead of Hosu et al. method | AestheticViT · AVA aesthetic score | SRCC: 0.756 -> 0.793 | Improved |
| Use ViT-L/16 21k instead of Chen et al. for classification | AestheticViT · AVA | accuracy (%): 83.20 -> 82.85 | Hurt |
| Use 32x32 patches (ViT-B/32) instead of 16x16 (ViT-B/16) (ablation) | ViT-B · AVA | SRCC: 0.759 -> 0.446 | Hurt |
| Scale ViT-B/16 to ViT-L/16 (ImageNet-21k) (ablation) | ViT · AVA | SRCC: 0.773 -> 0.793 | Improved |
| Pre-train ViT-B/16 on ImageNet-21k instead of ImageNet (ablation) | ViT-B/16 · AVA | SRCC: 0.759 -> 0.773 | Improved |
| Fine-tune full AestheticViT instead of linear probe on sentiment score (ablation) | ViT · RPCD | SRCC: 0.172 -> 0.250 | Improved |
| Fine-tune full AestheticViT instead of linear probe on sentiment score (ablation) | ViT · AVA | SRCC: 0.570 -> 0.544 | Hurt |
| Fine-tune full AestheticViT instead of linear probe on sentiment score (ablation) | ViT · AVA | accuracy (%): 76.43 -> 90.54 | Improved |
| Replace NIMA with AestheticViT for sentiment score | NIMA · RPCD | SRCC: 0.120 -> 0.250 | Improved |
| Train on RPCD vs PCCD for aesthetic captioning | BLIP · captioning | Bleu4: 0.011 -> 0.017 | Improved |
| Train on RPCD vs PCCD for aesthetic captioning | BLIP · captioning | CIDEr: 0.049 -> 0.048 | No clear effect |
Closest papers by meaning
- Bridging Cognitive Gap: Hierarchical Description Learning for Artistic Image Aesthetics Assessment (AAAI 2026 · similarity 0.90)
- Unlocking the Essence of Beauty: Advanced Aesthetic Reasoning with Relative-Absolute Policy Optimization (ICLR 2026 · similarity 0.87)
- Advancing Comprehensive Aesthetic Insight with Multi-Scale Text-Guided Self-Supervised Learning (AAAI 2025 · similarity 0.87)
- ArtiMuse: Fine-Grained Image Aesthetics Assessment with Joint Scoring and Expert-Level Understanding (CVPR 2026 · similarity 0.87)
- APDDv2: Aesthetics of Paintings and Drawings Dataset with Artist Labeled Scores and Comments (NeurIPS 2024 · similarity 0.87)
- Can Vision–Language Models Assess Graphic Design Aesthetics? A Benchmark, Evaluation, and Dataset Perspective. (ICLR 2026 · similarity 0.86)
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.
