Large language models (LLMs) are known for their exceptional performance in natural
language processing, making them highly effective in many human life-related or even job …

Large transformer models have achieved state-of-the-art results in numerous natural
language processing tasks. Among the pivotal components of the transformer architecture …

Large language models (LLMs) have made fundamental contributions over the last a few
years. To train an LLM, one needs to alternatingly runforward'computations andbackward' …

The Deep Leakage from Gradient (DLG) attack has emerged as a prevalent and highly
effective method for extracting sensitive training data by inspecting exchanged gradients …

Tensor decomposition is a fundamental method used in various areas to deal with high-
dimensional data.\emph {Tensor power method}(TPM) is one of the widely-used techniques …

Deploying Large Language Models (LLMs) in streaming applications that involve long
contexts, particularly for extended dialogues and text analysis, is of paramount importance …

In this paper, we consider a heavy inner product identification problem, which generalizes
the Light Bulb problem (1): Given two sets A⊂{-1,+1\}^d and B⊂{-1,+1\}^d with |A|=|B|=n, if …

Nuclear fusion has long attracted attention as a clean, efficient and sustainable form of
energy. However, complex physical design and engineering implementation are required to …