Research map / Label-efficient and robust learning / Recognition and long-tail
Predict and Constrain: Modeling Cardinality in Deep Structured Prediction
Read the paperCheck 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 |
|---|---|---|---|
| use Predict and Constrain with input-dependent cardinality prediction and projection | deep structured prediction model · Bibtex | macro F1: 44.7 -> 45.5 | Helped |
| use Predict and Constrain with input-dependent cardinality prediction and projection | deep structured prediction model · Bookmarks | macro F1: 37.1 -> 39.1 | Helped |
| use Predict and Constrain with input-dependent cardinality prediction and projection | deep structured prediction model · Delicious | macro F1: 37.8 -> 38.4 | Helped |
| replace sigmoid-based cardinality indicators with Predict and Constrain projection (ablation) | deep structured prediction model · Bibtex | macro F1: 42.0 -> 45.5 | Helped |
| replace differentiable fast exact projection with Predict and Constrain projection (ablation) | deep structured prediction model · Bibtex | macro F1: 42.1 -> 45.5 | Helped |
| replace sigmoid-based cardinality indicators with Predict and Constrain projection (ablation) | deep structured prediction model · Bookmarks | macro F1: 34.6 -> 39.1 | Helped |
| replace differentiable fast exact projection with Predict and Constrain projection (ablation) | deep structured prediction model · Bookmarks | macro F1: 36.0 -> 39.1 | Helped |
| replace sigmoid-based cardinality indicators with Predict and Constrain projection (ablation) | deep structured prediction model · Delicious | macro F1: 34.6 -> 38.4 | Helped |
| replace differentiable fast exact projection with Predict and Constrain projection (ablation) | deep structured prediction model · Delicious | macro F1: 34.2 -> 38.4 | Helped |
| predict cardinality from input rather than use random or constant cardinality | cardinality predictor on Delicious · Delicious | MSE: 74.9 -> 19.5; 26.9 -> 19.5 | Helped |
| use predicted rather than fixed cardinality | Predict and Constrain on Bibtex · Bibtex | macro F1 | Helped |
| use predicted rather than fixed cardinality | Predict and Constrain on Bookmarks · Bookmarks | macro F1 | Hurt |
Closest papers by meaning
- Structured Prediction Energy Networks (ICML 2016 · similarity 0.89)
- Joint Learning of Set Cardinality and State Distribution (AAAI 2018 · similarity 0.86)
- ADIOS: Architectures Deep In Output Space (ICML 2016 · similarity 0.85)
- Conformal Structured Prediction (ICLR 2025 · similarity 0.84)
- Learning with Structural Labels for Learning with Noisy Labels (CVPR 2024 · similarity 0.83)
- Taming the Sigmoid Bottleneck: Provably Argmaxable Sparse Multi-Label Classification (AAAI 2024 · 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.