针对多标签分类中标签表现不均的问题，提出公平偏好优化框架
FairPO: Robust Preference Optimization for Fair Multi-Label Learning
- 通过偏好损失聚焦困难标签，修正真实标签与混淆标签的排序错误
- 分组鲁棒优化动态平衡改进标签与基础标签的性能，减少偏差
- 支持无参考变体，适用于需要公平性的多标签场景
多标签分类常因标签间性能差异导致不公平。本文提出 FairPO 框架，结合基于偏好的损失与分组鲁棒优化，专门提升表现较差标签的性能。该框架将标签分为需重点改进的'优势标签集'和需保持基准性能的'非优势标签集'。对优势标签，采用受 DPO 启发的偏好损失，纠正真实标签与其混淆标签之间的排序错误；对非优势标签，通过约束目标保持性能稳定；并引入分组鲁棒偏好优化（GRPO）形式，自适应平衡两类目标以缓解偏差。同时验证了无需参考的对比（CPO）与简单（SimPO）偏好优化变体在该框架中的适用性。
原文摘要 · Abstract (English)
Multi-label classification (MLC) often suffers from performance disparities across labels. We propose \textbf{FairPO}, a framework combining preference-based loss and group-robust optimization to improve fairness by targeting underperforming labels. FairPO partitions labels into a \textit{privileged} set for targeted improvement and a \textit{non-privileged} set to maintain baseline performance. For privileged labels, a DPO-inspired preference loss addresses hard examples by correcting ranking errors between true labels and their confusing counterparts. A constrained objective maintains performance for non-privileged labels, while a Group Robust Preference Optimization (GRPO) formulation adaptively balances both objectives to mitigate bias. We also demonstrate FairPO's versatility with reference-free variants using Contrastive (CPO) and Simple (SimPO) Preference Optimization.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。