CAFE: Towards Compact, Adaptive, and Fast Embedding for Large-scale Recommendation Models

Open in new window