Abstract: This note investigates the logical matrix factorization with application to the topological structure analysis of Boolean networks. First, the concepts of both factorization and rank are ...

In this paper, we analyze two popular network compression techniques, i.e. filter pruning and low-rank decomposition, in a unified sense. By simply changing the way the sparsity regularization is ...

Palu is a KV-Cache compression framework that utilizes low-rank projection to compress the hidden dimension of KV-Cache, thereby reducing memory footprint and increasing speed. Palu is a pioneer ...

Mathematicians love a good puzzle. Even something as abstract as multiplying matrices (two-dimensional tables of numbers) can feel like a game when you try to find the most efficient way to do it.

An important paradigm of natural language processing consists of large-scale pre-training on general domain data and adaptation to particular tasks or domains. As we pre-train larger models, full fine ...

Extensive clinical and biomedical studies have shown that microbiome plays a prominent role in human health. Identifying potential microbe–disease associations (MDAs) can help reveal the pathological ...

As artificial intelligence (AI) applications become ubiquitous in medical care, autonomous driving, robotics, and other fields, accuracy requirements and neural network complexity increase in tandem, ...

Molecular simulations of proteins are often interpreted using Markov state models (MSMs), in which each protein configuration is assigned to a global state. As we explore larger and more complex ...