Research map / Label-efficient and robust learning / Noisy labels
Learning to Aggregate Ordinal Labels by Maximizing Separating Width
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 |
|---|---|---|---|
| compare with G-CrowdSVM | separating-width aggregation · Bird | l0 error rate (%): 10.37±0.41 -> 9.25±0.17 | Helped |
| compare with Opt-DS | separating-width aggregation · TREC | l0 error rate (%): 29.80 -> 29.30±0.11 | Helped |
| compare with Opt-DS and MV-DS | separating-width aggregation · RTE | l0 error rate (%): 7.12 -> 7.00±0.29 | Helped |
| set lambda1=lambda2=0 to degenerate to MV-DS (no worker/item priors) | separating-width aggregation · binary benchmarks | l0 error rate (%): 7.12 -> 7.00±0.29 | Hurt |
| optimize separating width and update priors vs G-CrowdSVM | separating-width aggregation · Web | l0 error: 0.0799±0.0026 -> 0.0322±0.0013 | Helped |
| same comparison with G-CrowdSVM | separating-width aggregation · Web | l2 error: 0.3629±0.0044 -> 0.2153±0.0019 | Helped |
| same comparison with Entropy(O) | separating-width aggregation · Web | l1 error: 0.1173 -> 0.0369±0.0032 | Helped |
| compare with G-CrowdSVM | separating-width aggregation · Age | l0 error: 0.3298±0.0036 -> 0.3210±0.0025 | Helped |
| compare with majority voting | separating-width aggregation · Age | l2 error: 0.7297 -> 0.6192±0.0047 | Helped |
| compare with Eig-Ratio | separating-width aggregation · RTE | l0 error rate (%): 9.00 -> 7.00±0.29 | Helped |
| report l1 alongside l0 for ordinal labels | separating-width aggregation · Web | l1 vs l0 error: 0.0322±0.0013 -> 0.0369±0.0032 | No clear effect |
Closest papers by meaning
- Unbiased Multi-Label Learning from Crowdsourced Annotations (ICML 2024 · similarity 0.86)
- Label Aggregation for Composite Crowd Tasks by Worker Ability Constraint Satisfaction (AAAI 2025 · similarity 0.85)
- Label Distribution Propagation-based Label Completion for Crowdsourcing (ICML 2025 · similarity 0.85)
- Learning from Noisy Labels via Conditional Distributionally Robust Optimization (NeurIPS 2024 · similarity 0.85)
- TLLC: Transfer Learning-based Label Completion for Crowdsourcing (ICML 2025 · similarity 0.84)
- Efficient Online Crowdsourcing with Complex Annotations (AAAI 2024 · 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.