Research map / Label-efficient and robust learning / Recognition and long-tail

ComRank: Ranking Loss for Multi-Label Complementary Label Learning

NeurIPS 2025 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 0 citations (OpenAlex)

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes ComRank, a ranking-loss approach for multi-label complementary label learning that avoids the uniform-distribution assumption behind unbiased risk estimators.

Problem

Multi-label complementary label learning (MLCLL), where only irrelevant (complementary) labels are available instead of relevant labels; existing unbiased risk estimators assume complementary labels follow a uniform distribution, which fails in real-world settings.

Approach

Replaces the unbiased risk estimator, which assumes uniformly distributed complementary labels, with a ranking loss that ranks relevant labels above complementary ones, so the method does not rely on that distributional assumption.

Experimental setup

Baselines: unbiased risk estimator methods for MLCLL

Builds on: unbiased risk estimator complementary label learning

Closest papers by meaning

• Multi-Label Learning with Stronger Consistency Guarantees (NeurIPS 2024 · similarity 0.89)
• Learning with Complementary Labels Revisited: The Selected-Completely-at-Random Setting Is More Practical (ICML 2024 · similarity 0.88)
• Semi-supervised Multi-label Learning with Balanced Binary Angular Margin Loss (NeurIPS 2024 · similarity 0.87)
• Bipartite Ranking From Multiple Labels: On Loss Versus Label Aggregation (ICML 2025 · similarity 0.86)
• Rebalancing Multi-Label Class-Incremental Learning (AAAI 2025 · similarity 0.86)
• Addressing Multi-Label Learning with Partial Labels: From Sample Selection to Label Selection (AAAI 2025 · similarity 0.86)