English(EN)ICEGR: An Intent-Coherent End-to-End Generative Retrieval Framework for E-commerce Search
电子商务搜索因新的生成检索研究而得到提升 · 跟踪 4 个来源
作者PulseAugur 编辑部·[6 个来源]·
arXiv 上发表的四篇研究论文提出了电子商务搜索中生成检索的新方法。这些方法旨在通过联合训练嵌入模型和码本、整合产品集群信息以及优化相关性指标来提高产品搜索的准确性和用户参与度。提出的框架解决了诸如误差累积、嘈杂的用户参与信号以及需要更好的查询意图一致性等问题，其中一些框架在生产环境中展示了检索指标和业务成果的显著改进。
AI
arXiv:2608.30606v1 Announce Type: cross Abstract: With the development of large language models (LLMs), generative retrieval is becoming increasingly important in e-commerce scenarios. Current mainstream approaches typically use a two-stage training strategy: first train a produc…
With the development of large language models (LLMs), generative retrieval is becoming increasingly important in e-commerce scenarios. Current mainstream approaches typically use a two-stage training strategy: first train a product embedding model, and then learn a codebook that …
With the development of large language models (LLMs), generative retrieval is becoming increasingly important in e-commerce scenarios. Current mainstream approaches typically use a two-stage training strategy: first train a product embedding model, and then learn a codebook that …
arXiv:2602.17654v2 Announce Type: replace-cross Abstract: Embedding-based retrieval (EBR) for large-scale e-commerce search faces three intertwined challenges: graded (non-binary) relevance where engagement signals are noisy and intent-varying while business relevance guidelines …
Generative Retrieval (GR) is promising for e-commerce search, yet existing methods struggle to maintain query-intent consistency throughout the training pipeline. First, semantic ID (SID) construction based on static product information limits the ability of SIDs to encode produc…
Generative Retrieval (GR) is promising for e-commerce search, yet existing methods struggle to maintain query-intent consistency throughout the training pipeline. First, semantic ID (SID) construction based on static product information limits the ability of SIDs to encode produc…
