CoRe: A Continuously Reward-Finetuned LLM Query Rewriter for Multi-Stage Context-Aware Relevance in Web-Scale Video Search
- 类型:arxiv
- 标识:2606.14127
- 链接:http://arxiv.org/abs/2606.14127v1
- 主分类:engineering
- 形态:application
- 被引:0
- 被引来源:Semantic Scholar + OpenAlex
- S2被引:0
- OpenAlex被引:0
- 影响力被引:0
- TLDR:CoRe (Context Relevance) is presented, such a system, redeployed weekly for over five months in a major short-video search engine, using the deployed multimodal relevance model as its source and a multiplicative ratio form mirroring the production fusion algebra to close the simulation-production gap.
- OpenAlex ID:W7164848026
- OpenAlex DOI:10.48550/arxiv.2606.14127
- DOI:10.48550/arxiv.2606.14127
- DOI来源:OpenAlex
- 开放获取:green
- 开放获取链接:https://doi.org/10.48550/arxiv.2606.14127
- OpenAlex更新:2026-07-19
- 副分类:multimodal
- 待LLM分类:否
- 标题中文:CoRe:面向 Web 规模视频搜索中多阶段上下文感知相关性的连续奖励微调 LLM 查询改写器
- TLDR中文:CoRe(Context Relevance)被提出。该系统在一个大型短视频搜索引擎中每周重新部署,持续运行超过五个月,使用已部署的多模态相关性模型作为源,并以镜像生产融合代数的乘性比值形式来缩小仿真与生产之间的差距。
- 来源文件:
- /inbox/tom/_candidates/2026-06-17-rag-retrieval-reranking-candidates.json
- [S2 enrich]
- [OpenAlex backfill]