DREAM proposes intent-aware tokenization, frozen-model evaluation, and dynamic beams to refine early SID assignments and improve cold-start performance in generative recommenders on Amazon benchmarks.
Title resolution pending
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
citing papers explorer
-
DREAM: Dynamic Refinement of Early Assignment Mappings
DREAM proposes intent-aware tokenization, frozen-model evaluation, and dynamic beams to refine early SID assignments and improve cold-start performance in generative recommenders on Amazon benchmarks.
- FLUID: From Ephemeral IDs to Multimodal Semantic Codes for Industrial-Scale Livestreaming Recommendation