Analog bits: Generating discrete data using diffusion models with self-conditioning

T Chen, R Zhang, G Hinton - arXiv preprint arXiv:2208.04202, 2022 - arxiv.org
We present Bit Diffusion: a simple and generic approach for generating discrete data with
continuous state and continuous time diffusion models. The main idea behind our approach …

Learning vector-quantized item representation for transferable sequential recommenders

Y Hou, Z He, J McAuley, WX Zhao - … of the ACM Web Conference 2023, 2023 - dl.acm.org
Recently, the generality of natural language text has been leveraged to develop transferable
recommender systems. The basic idea is to employ pre-trained language models (PLM) to …

Lightweight self-attentive sequential recommendation

Y Li, T Chen, PF Zhang, H Yin - Proceedings of the 30th ACM …, 2021 - dl.acm.org
Modern deep neural networks (DNNs) have greatly facilitated the development of sequential
recommender systems by achieving state-of-the-art recommendation performance on …

Compositional embeddings using complementary partitions for memory-efficient recommendation systems

HJM Shi, D Mudigere, M Naumov, J Yang - Proceedings of the 26th ACM …, 2020 - dl.acm.org
Modern deep learning-based recommendation systems exploit hundreds to thousands of
different categorical features, each with millions of different categories ranging from clicks to …

Compressing word embeddings via deep compositional code learning

R Shu, H Nakayama - arXiv preprint arXiv:1711.01068, 2017 - arxiv.org
Natural language processing (NLP) models often require a massive number of parameters
for word embeddings, resulting in a large storage or memory footprint. Deploying neural …

Efficient on-device session-based recommendation

X Xia, J Yu, Q Wang, C Yang, NQV Hung… - ACM Transactions on …, 2023 - dl.acm.org
On-device session-based recommendation systems have been achieving increasing
attention on account of the low energy/resource consumption and privacy protection while …

Lightrec: A memory and search-efficient recommender system

D Lian, H Wang, Z Liu, J Lian, E Chen… - Proceedings of The Web …, 2020 - dl.acm.org
Deep recommender systems have achieved remarkable improvements in recent years.
Despite its superior ranking precision, the running efficiency and memory consumption turn …

Knowledge distillation for high dimensional search index

Z Lu, J Chen, D Lian, Z Zhang… - Advances in Neural …, 2024 - proceedings.neurips.cc
Lightweight compressed models are prevalent in Approximate Nearest Neighbor Search
(ANNS) and Maximum Inner Product Search (MIPS) owing to their superiority of retrieval …

Differentiable product quantization for end-to-end embedding compression

T Chen, L Li, Y Sun - International Conference on Machine …, 2020 - proceedings.mlr.press
Embedding layers are commonly used to map discrete symbols into continuous embedding
vectors that reflect their semantic meanings. Despite their effectiveness, the number of …

A novel dependency-oriented mixed-attribute data classification method

YL He, GL Ou, P Fournier-Viger, JZ Huang… - Expert Systems with …, 2022 - Elsevier
How to design an efficient method to handle mixed-attribute data classification (MADC)
problems has become a hot topic in data mining and machine learning. Current MADC …