Research map / Learning theory and algorithms / Prediction and decision losses
Online Conformal Selection with Accept-to-Reject Changes
Kangdao Liu, Huajun Xi, Chi Man Vong, Hongxin Wei
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 OCS-ARC instead of online Bonferroni (OB) baseline | OCS-ARC · Synthetic regression (gradient boosting) | power | Improved |
| use OCS-ARC instead of OB baseline | OCS-ARC · Synthetic regression (gradient boosting) | FDR_t | Improved |
| use OCS-ARC instead of OB baseline | OCS-ARC · Recruitment screening (Kaggle) | power | Improved |
| use CLIP score instead of RES score (ablation) | OCS-ARC · Synthetic regression | power | Improved |
| increase noise level in the data (ablation) | OCS-ARC · Synthetic regression | power | Hurt |
| increase decay coefficient r to 0.999 (ablation) | OCS-ARC · Synthetic regression Setting 1 | power | Hurt |
| vary decay coefficient r over 0.99 to 0.999 (ablation) | OCS-ARC · Synthetic regression Setting 1 | FDR_t | No clear effect |
| enlarge calibration set (ablation) | OCS-ARC · Synthetic regression Setting 1 | FDR_t | Mixed |
| apply OCS-ARC with LLaMA-2-13B-chat alignment scores | OCS-ARC · TriviaQA | FDP_t | Improved |
Closest papers by meaning
- Conformalized Multiple Testing after Data-dependent Selection (NeurIPS 2024 · similarity 0.89)
- Multivariate Conformal Selection (ICML 2025 · similarity 0.89)
- Online Localized Conformal Prediction (NeurIPS 2026 · similarity 0.85)
- Valid Selection among Conformal Sets (NeurIPS 2025 · similarity 0.85)
- Real-Time Selection Under General Constraints via Predictive Inference (NeurIPS 2024 · similarity 0.84)
- On the Design Space of Discrete Diffusion Online Adaptation for Molecular Optimization (NeurIPS 2026 · similarity 0.84)
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.
