发表机构
Tsinghua University; University of Science and Technology of China（清华大学; 中国科学技术大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对扩散推荐模型因生成动态导致流行度偏差自我强化的问题，提出即插即用的公平感知框架 FairDiff，通过流行度条件引导与语义校准模块，在保持最先进性能的同时有效缓解马太效应。
AI 中文摘要
虽然“马太效应”和过滤气泡被广泛认为是推荐系统中结果层面的偏差，但我们揭示出扩散推荐模型（DRMs）通过其生成动态独特地加剧了这一问题。DRMs 不仅仅是继承了数据不平衡，而是触发了流行度偏差的自我强化放大。我们识别出这一现象由两种复合机制驱动。首先，虽然优化损失在各类推荐器中普遍由高频项目主导，但 DRMs 在生成过程中遭受独特的结构先验不匹配。由于长尾数据的前向终端分布显著偏离标准高斯先验，反向采样轨迹固有地向高密度流行项目坍缩，从根本上抑制了利基项目的生成。为了打破这一自我强化循环，我们提出了 FairDiff，一个即插即用的公平感知扩散框架。为了克服流行度主导的损失，我们引入了流行度条件引导（PCG）。PCG 不改变训练目标，而是作为推理时的分布重加权机制，通过数学方式重塑基于分数的梯度场，以惩罚高流行度区域并将轨迹引向利基语义。此外，我们设计了一个语义校准（SC）模块来弥合先验不匹配，通过一步最优传输对齐前向和反向分布。综合评估表明，FairDiff 在有效缓解自我强化马太效应的同时实现了最先进的性能，凸显了其作为 DRMs 通用框架的价值。
英文摘要
While the "Matthew Effect" and filter bubbles are widely recognized outcome-level biases in recommender systems, we reveal that Diffusion Recommender Models (DRMs) uniquely compound this issue through their generative dynamics. Rather than merely inheriting data imbalances, DRMs trigger a self-reinforcing amplification of popularity bias. We identify that this phenomenon is driven by two compounding mechanisms. First, while optimization loss is universally dominated by high-frequency items across recommenders, DRMs suffer from a unique structural prior mismatch during generation. Because the forward terminal distribution of long-tailed data deviates significantly from the standard Gaussian prior, reverse sampling trajectories inherently collapse toward high-density popular items, fundamentally suppressing niche item generation. To dismantle this self-reinforcing loop, we propose FairDiff, a plug-and-play fairness-aware diffusion framework. To overcome the popularity-dominated loss, we introduce Popularity Condition Guidance (PCG). Rather than altering the training objective, PCG acts as an inference-time distributional reweighting mechanism, mathematically reshaping the score-based gradient field to penalize high-popularity regions and guide trajectories toward niche semantics. Furthermore, we design a Semantic Calibration (SC) Module to bridge the prior mismatch, aligning the forward and reverse distributions via one-step optimal transport. Comprehensive evaluations demonstrate that FairDiff achieves state-of-the-art performance while effectively mitigating the self-reinforcing Matthew Effect, highlighting its value as a general framework for DRMs.
