相关论文: Hurwitz Quaternion Multiplicative Quantization for…

To enable broader deployment of Large Language Models (LLMs), it is essential to identify the best-performing model under strict memory constraints. We present AMQ, Automated Mixed-Precision Weight-Only Quantization, a framework that…

Quantization emerges as one of the most promising compression technologies for deploying efficient large models for various real time application in recent years. Considering that the storage and IO of weights take up the vast majority of…

AlphaFold has achieved groundbreaking advancements in protein structure prediction, exerting profound influence across biology, medicine, and drug discovery. However, its reliance on multiple sequence alignment (MSA) is inherently…

This paper presents a comprehensive analysis of quantization techniques for optimizing Large Language Models (LLMs), specifically focusing on Post-Training Quantization (PTQ) and Quantization-Aware Training (QAT). Through empirical…

Large language models have transformed the comprehension and generation of natural language tasks, but they come with substantial memory and computational requirements. Quantization techniques have emerged as a promising avenue for…

Memory consumption of the Key-Value (KV) cache represents a major bottleneck for efficient large language model inference. While attention-score-based KV cache pruning shows promise, it faces critical practical limitations: attention scores…

Vision-language models (VLMs) show remarkable performance in multimodal tasks. However, excessively long multimodal inputs lead to oversized Key-Value (KV) caches, resulting in significant memory consumption and I/O bottlenecks. Previous KV…

We introduce the Holographic Knowledge Manifold (HKM), a four-phase pipeline that achieves zero catastrophic forgetting in AI knowledge representation while maintaining minimal memory growth and high efficiency. Leveraging fractal…

The Worldvolume Hybrid Monte Carlo (WV-HMC) method [arXiv:2012.08468] is an efficient algorithm for addressing the numerical sign problem at moderate computational cost. It mitigates the sign problem while avoiding the ergodicity issues…

Large Language Models (LLMs), despite their recent impressive accomplishments, are notably cost-prohibitive to deploy, particularly for applications involving long-content generation, such as dialogue systems and story writing. Often, a…

Developing efficient quantum computing algorithms is essential for tackling computationally challenging problems across various fields. This paper presents a novel quantum algorithm, XZ24, for efficiently computing the eigen-energy spectra…

Large Language Models (LLMs) excel in NLP, but their demands hinder their widespread deployment. While Quantization-Aware Training (QAT) offers a solution, its extensive training costs make Post-Training Quantization (PTQ) a more practical…

As the size of large language models (LLMs) continues to grow, model compression without sacrificing accuracy has become a crucial challenge for deployment. While some quantization methods, such as GPTQ, have made progress in achieving…

KV-cache quantizers usually optimize storage-space reconstruction, even though attention reads keys through logits and values through attention-weighted readout. We argue that persistent cache error should be measured in model-visible…

Processing qubit Hamiltonians derived from electronic-structure problems can become classically prohibitive because many downstream manipulations still rely on dense operator constructions whose cost grows exponentially with qubit number.…

Post-training quantization (PTQ) of large language models (LLMs) to extremely low bit-widths remains challenging due to the fundamental trade-off between computational efficiency and representational capacity. While existing ultra-low-bit…

The conventional approach to the fronthaul design for cell-free massive MIMO system follows the compress-and-precode (CP) paradigm. Accordingly, encoded bits and precoding coefficients are shared by the distributed unit (DU) on the…

We report an efficient program for computing the eigenvalues and symmetry-adapted eigenvectors of very large quaternionic (or Hermitian skew-Hamiltonian) matrices, using which structure-preserving diagonalization of matrices of dimension N…

This paper presents key enhancements to our previous work~\cite{naghmouchi2024mixed} on a hybrid Benders decomposition (HBD) framework for solving mixed integer linear programs (MILPs). In our approach, the master problem is reformulated as…

Speculative decoding and quantization effectively accelerate memory-bound inference of large language models. Speculative decoding mitigates the memory bandwidth bottleneck by verifying multiple tokens within a single forward pass, which…