ReMem: Rethinking Perception and Memory in Long-Context Recommendation Agents
A novel recommendation agent framework, termed as ReMem, that combines OCR-based multimodal perception with time-evolving dynamic memory, which allows the agent to preserve evolving user preferences without relying on external memory modules or disrupting the standard autoregressive generation process.
Hao-Hao Qu, Yong-Cheng Jing, Chun-Hin Chan et al.
· 0 citations