检验多模态推荐中用户专属权重是否真有效，发现全局权重已够用。
Is Personalized Modality Weighting Actually Personalized? A Controlled Audit of Per-User Weighting Claims in Multimodal Recommenders
- 用统一框架对比六种权重方法，控制变量验证用户专属性
- 全局权重在三数据集上已接近全部增益（最高+3.6pp），用户专属无持续提升
- 建议报告真实增益与乱序对照双指标，作为个性化声明最低标准
多模态推荐系统在数十亿用户中部署了用户专属模态权重，声称能提升排序效果。但现有评估未区分用户专属信号与全局权重加模型容量的影响。本文采用双对比审计法，将六种实现简化为同一协同基础模型，分别测量真实增益（real-GM）与可辨识性缺口（real-shuf）。在三个短视频数据集上，单一全局权重已带来几乎全部内容增益（较无模态基线+1.9/+3.6/+3.5pp，p < .001）。引入用户专属权重后，无一方法在所有数据集和指标上胜出，少数正向差距极小（≤0.9pp）且方向反转。乱序对照显示，real-shuf可达全局增益的128%，说明其无法充分检验个性化。溯源发现，门控机制依赖共享协同嵌入，切断输入后real-shuf近归零，而实用性结论不变。剂量反应实验验证了检测工具的有效性（AUROC从0.57升至0.89，0.64升至1.00）。所有结果在第四个跨域电商数据集上复现。建议未来报告real-GM与real-shuf双指标作为个性化证据的最低标准。
原文摘要 · Abstract (English)
Per-user modality weighting is deployed at billion-user scale in multimodal recommenders, through user modality-strength vectors, attention gates, meta-weight hypernetworks, and low-rank guided weights, each claiming a ranking gain from user-specific modality preference. Yet, to our knowledge, prior evaluations do not isolate a genuinely user-specific signal from a global modality weight plus model capacity. We audit this family with a two-contrast audit principle, reducing six implementations onto one shared collaborative backbone and measuring a utility gap (real-GM) against a single global modality weight and an identifiability gap (real-shuf) against an eval-time permutation of the user-weight binding. Across three independent short-video corpora, a single global weight already delivers nearly all of the content gain (+1.9/+3.6/+3.5pp over a no-modality baseline, p < .001). Making the weight per-user adds no consistent utility: no implementation wins on all corpora and metrics, and the few positive gaps are small (<=0.9pp) and flip. The shuffle control is necessary but not sufficient, since real-shuf reaches +128% of the content gain for heads that simultaneously lose to the global weight. We trace this dissociation to gates reading the shared collaborative embedding: decoupling the gate input collapses the inflated real-shuf to near zero while the utility conclusion stands. A monotone signal-implant dose-response (capture AUROC rising from 0.57 to 0.89 and from 0.64 to 1.00) verifies the harness would detect user-specific structure if present, and every finding replicates on a fourth, cross-domain e-commerce corpus. We propose reporting real-GM alongside real-shuf as a minimum evidentiary standard for personalization claims.
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。