Research map / Label-efficient and robust learning / Test-time adaptation
Resource Constrained 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 adaptive part-by-part feature selection instead of uniform feature allocation | Proposed OCR one-shot policy · OCR | running time at comparable performance: up to 50 percent | Helped |
| Use complex features with part-by-part selection instead of uniform feature templates | OCR policy · OCR | feature efficiency | Helped |
| Use simple features instead of complex features in the low-budget region | OCR policy · OCR, low-budget region | total running time | Helped |
| Use part-by-part feature selection instead of global feature selection | Proposed structured prediction policy · Structured prediction | policy performance | Helped |
| Use one-shot feature selection instead of anytime selection | Dependency parser · Penn Treebank | unlabeled attachment accuracy across budget levels | Helped |
| Use a myopic confidence-based policy instead of the uniform policy | Dependency parser · Penn Treebank | overall performance: +4.5 seconds of extra time | Hurt |
| Use greedy trajectory search instead of parsimonious search (ablation) | One-shot dependency parser · Penn Treebank, low-budget regions | unlabeled attachment accuracy | Helped |
| Add importance weights to the greedy policy (ablation) | Greedy dependency parsing policy · Penn Treebank | policy performance | Helped |
| Apply length dictionary filtering (ablation) | Dependency parser · Penn Treebank section 23 | UAS: 89.7%; losing only 1% | Mixed |
| Concentrate expensive feature extraction near the root and other low-depth words (ablation) | Dependency parsing policy · Penn Treebank | accuracy gains from feature allocation | Helped |
Closest papers by meaning
- A Smoother Way to Train Structured Prediction Models (NeurIPS 2018 · similarity 0.85)
- On Structured Prediction Theory with Calibrated Convex Surrogate Losses (NeurIPS 2017 · similarity 0.85)
- A Consistent Regularization Approach for Structured Prediction (NeurIPS 2016 · similarity 0.84)
- Bandit and Delayed Feedback in Online Structured Prediction (NeurIPS 2025 · similarity 0.83)
- Efficient Adaptive Data Acquisition via Pretrained Belief Representations (NeurIPS 2026 · similarity 0.82)
- Ranked Entropy Minimization for Continual Test-Time Adaptation (ICML 2025 · similarity 0.82)
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.