LEAF: Lightweight, Efficient, Adaptive and Flexible Embedding for Large-Scale Recommendation Models
Open Access
- 22 September 2025
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 217-225
Abstract
No abstract availableFunding Information
- Defense Advanced Research Projects Agency (DARPA) (HR001120C0088)
- NSF (National Science Foundation) (2224319)
- REAL@USC-Meta center (gift)
- Broadcom (gift)
This publication has 20 references indexed in Scilit:
- AutoEmb: Automated Embedding Dimensionality Search in Streaming Recommendations2021
- Mixed Dimension Embeddings with Application to Memory-Efficient Recommendation Systems2021
- The Architectural Implications of Facebook's DNN-Based Personalized Recommendation2020
- Wide & Deep Learning for Recommender Systems2016
- Deep Neural Networks for YouTube Recommendations2016
- Beating CountSketch for heavy hitters in insertion streams2016
- Space-optimal heavy hitters with strong error boundsACM Transactions on Database Systems, 2010
- Methods for finding frequent items in data streamsThe VLDB Journal, 2009
- Feature hashing for large scale multitask learning2009
- An improved data stream summary: the count-min sketch and its applicationsJournal of Algorithms, 2005