作者位次 #2/3ECCV 2026Poster
Personalized Reward Modeling for Text-to-Image Generation
Jeongeun Lee · Ryang Heo · Dongha Lee
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Jeongeun Lee · Ryang Heo · Dongha Lee
Sangam Lee · Ryang Heo · SeongKu Kang · Susik Yoon · Jinyoung Yeo · Dongha Lee
Generative retrieval directly decode a document identifier (i.e., docid) in response to a query, making it impossible to provide users with explanations as an answer for “why is this document retrieved?”. To address this limitation, we propose Hierarchical Category Path-Enhanced Generative Retrie...