Research map / Vision-language and multimodal / Human and embodied modeling
Some Like It Hot - Visual Guidance for Preference Prediction
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Add visual regularization to matrix factorization (ablation) | MF for Hot-or-Not female queries · Hot-or-Not, 10 Ratings | accuracy: 72.92% -> 78.68% | Helped |
| Add visual regularization to matrix factorization (ablation) | MF for Hot-or-Not female queries · Hot-or-Not, visual only | accuracy: 75.90% -> 75.92% | No clear effect |
| Add visual regularization to matrix factorization (ablation) | MF for Hot-or-Not female queries · Hot-or-Not, 100 Ratings | accuracy: 79.82% -> 81.82% | Helped |
| Add visual regularization to matrix factorization (ablation) | MF for Hot-or-Not female queries · Hot-or-Not, Full History | accuracy: 83.62% -> 83.64% | No clear effect |
| Add visual regularization to matrix factorization (ablation) | MF for MovieLens · MovieLens 10M, visual only | MAE: 0.824 -> 0.813 | Helped |
| Add visual regularization to matrix factorization (ablation) | MF for MovieLens · MovieLens 10M, 10 Ratings | MAE: 1.031 -> 0.872 | Helped |
| Add visual regularization to matrix factorization (ablation) | MF for MovieLens · MovieLens 10M, 100 Ratings | MAE: 0.740 -> 0.780 | Hurt |
| Add visual regularization to matrix factorization (ablation) | MF for MovieLens · MovieLens 10M, Full History | MAE: 0.696 -> 0.696 | No clear effect |
| Use VGG-16 deep features with SVR instead of the DLA baseline | Age estimation · MORPH 2 | MAE: 4.77 -> 3.45 | Helped |
| Use VGG-16 visual features for facial beauty estimation | Facial beauty prediction · Gray dataset, without alignment | Pearson correlation: 0.425 -> 0.470 | Helped |
Closest papers by meaning
- PreferThinker: Reasoning-based Personalized Image Preference Assessment (ICLR 2026 · similarity 0.86)
- Learning Multi-Dimensional Human Preference for Text-to-Image Generation (CVPR 2024 · similarity 0.85)
- Diffusion Model as a Noise-Aware Latent Reward Model for Step-Level Preference Optimization (NeurIPS 2025 · similarity 0.84)
- UniAR: A Unified model for predicting human Attention and Responses on visual content (NeurIPS 2024 · similarity 0.84)
- Aligning Vision Models with Human Aesthetics in Retrieval: Benchmarks and Algorithms (NeurIPS 2024 · similarity 0.83)
- Personalized Preference Fine-tuning of Diffusion Models (CVPR 2025 · similarity 0.83)
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.