1
CoRe: A Continuously Reward-Finetuned LLM Query Rewriter for Multi-Stage Context-Aware Relevance in Web-Scale Video Search
将大规模视频搜索中的查询改写难题转化为连续奖励微调任务,实现多阶段上下文感知的相关性提升。
arXiv:2606.14127v1 Announce Type: cross Abstract: LLM-based query rewriters in production face a tension: the training reward must reflect how the rew…