arXiv：cs.AI· Yingtao Ren, Aolin Li, Cheng Cheng, Huiyun Hu, Huasheng Liang, Baoxing Xie, Jun Yuan, Haijun Wu, Chenxing Wang·· 16 小时前
WeFair：面向微信短视频搜索个性化查询改写的人类反馈保真对齐
WeFair: Fidelity-Aware Alignment from Human Feedback for Personalized Query Rewriting in Short-Video Search
AI 导读
微信短视频搜索提出 WeFair 查询改写方法，将真实用户行为信号划分为演示、偏好、验证三个层级，先用搜索会话中的人类改写数据做监督课程学习，再用 DPO 对齐用户后续消费行为，最后以严格验证奖励结合 GRPO 做在线策略优化。线上 A/B 测试显示，有效点击查询率（VV>10s）相对提升 1.576%，查询改写率相对下降 1.731%。
正文
Abstract:Short-video search requires compact query rewriters that infer personalized intent from brief, ambiguous queries under strict latency constraints. The challenge is to decide when personalization is warranted and generate a reformulation that preserves the user's immediate information need. A rewriter optimized for fluency or engagement may still introduce unsupported details and redirect the search toward a different intent. Existing methods often underutilize users' actual reformulations and behavioral traces, relying instead on synthetic reasoning and indirect behavioral rewards that may produce more engaging candidates but cause intent drift. To build an effective rewriter that remains faithful to user intent, we propose WeFair, Fidelity-Aware Alignment from Human Feedback for Personalized Query Rewriting in WeChat short-video search. WeFair assigns real user behavioral signals to a progression from demonstration to preference to verification. We first mine actual human reformulations from search sessions, then apply multidimensional LLM screening and contextual verification to filter unsupported transitions without extensive manual annotation. During training, we begin with a supervised curriculum that progressively builds rewriting capabilities from large-scale human personalized query demonstrations, and then use direct preference optimization to align with users' subsequent consumption. In the final on-policy optimization stage, we construct a strict verification reward based on actual human reformulations to further boost performance with group relative policy optimization. Further analysis shows that auxiliary rewards can activate more rollout groups in GRPO but reduce intent fidelity. Online A/B testing in WeChat short-video search shows a 1.576% relative increase in effective clicked-query rate (VV>10s) and a 1.731% relative reduction in query reformulation rate.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.16927 [cs.AI] |
| (or arXiv:2610.16927v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.16927 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Yingtao Ren [view email]
[v1]
Mon, 12 Oct 2026 15:10:01 UTC (2,342 KB)
来源：arXiv：cs.AI · arxiv.org