Research map / Time series / Foundation models for time series
POP: Mining POtential Performance of New Fashion Products via Webly Cross-Modal Query Expansion
Christian Joppi, Geri Skenderi, Marco Cristani
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 |
|---|---|---|---|
| add POP signal in place of Google Trends | Concat MM RNN · VISUELLE (first order setup) | WAPE: 64.11 -> 58.09 | Improved |
| add POP signal in place of Google Trends | GTM Transformer · VISUELLE (release setup) | WAPE: 55.24 -> 52.39 | Improved |
| add POP signal in place of Google Trends | Gradient Boosting · VISUELLE (release setup) | WAPE: 63.52 -> 63.38 | No clear effect |
| replace No Signal with Google Trends | Residual MM RNN · VISUELLE (release setup) | WAPE: 64.26 -> 68.46 | Hurt |
| add time-dependent query expansion (ablation) | POP · VISUELLE (first order setup) | WAPE: 54.47 -> 53.41 | Improved |
| add time-dependent query expansion (ablation) | POP · VISUELLE (release setup) | WAPE: 53.12 -> 52.39 | Improved |
| use correctly aligned past instead of one-year misaligned past (ablation) | POP · VISUELLE (first order setup) | WAPE: 53.63 -> 53.41 | Improved |
| add fine-tuning on web data (ablation) | POP · VISUELLE (release setup) | WAPE: 53.03 -> 52.81 | Improved |
| add confident-learning pruning of noisy labels (ablation) | POP · VISUELLE (release setup) | WAPE: 52.81 -> 52.39 | Improved |
| confident learning versus SELFIE (ablation) | POP · VISUELLE (first order setup) | WAPE: 53.51 -> 53.41 | Improved |
| use only positive-image similarity instead of negative signal (ablation) | POP · VISUELLE (release setup) | WAPE: 53.90 -> 53.41 | Improved |
| add negative signal as second channel to POP (ablation) | POP · VISUELLE (first order setup) | WAPE: 52.39 -> 52.97 | Hurt |
Closest papers by meaning
- Fashion Forward: Forecasting Visual Style in Fashion (ICCV 2017 · similarity 0.89)
- Context-Aware Visual Compatibility Prediction (CVPR 2019 · similarity 0.85)
- Fashion IQ: A New Dataset Towards Retrieving Images by Natural Language Feedback (CVPR 2021 · similarity 0.85)
- MMTryon: Multi-Modal Multi-Reference Control for High-Quality Fashion Generation (TPAMI 2026 · similarity 0.85)
- From Paris to Berlin: Discovering Fashion Style Influences Around the World (CVPR 2020 · similarity 0.84)
- StyleTailor: Towards Personalized Fashion Styling via Hierarchical Negative Feedback (AAAI 2026 · 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.
