arXiv:2606. 04050v1 Announce Type: cross Abstract: Existing quantization methods are fundamentally limited by rigid, integer-based bit-widths (e.
arXiv:2602. 06694v3 Announce Type: replace Abstract: Weight-only quantization has become a standard approach for efficiently serving large language models (LLMs).
arXiv:2608. 08081v1 Announce Type: cross Abstract: Large mixture-of-experts (MoE) language models with 26--120 billion parameters exceed the memory capacity of consumer devices through three simultaneous pressures: resident weight matrices, key-value (KV) cache state that grows linearly with context, and dozens of expert sublayers that must be paged on demand.
arXiv:2607. 14618v1 Announce Type: new Abstract: CPUs are the most universal target for on-device LLM inference, but existing low-bit quantization methods offer either coarse operating points or fine-grained mixed precision that is difficult to execute efficiently on CPUs.
arXiv:2606. 00079v1 Announce Type: cross Abstract: Mixture-of-Experts (MoE) large language models reduce per-token computation through sparse expert activation, but their deployment remains memory-intensive because all expert weights must be kept resident in memory.
arXiv:2605. 08692v2 Announce Type: replace Abstract: Post-training weight-only quantization to 4 bits is widely used to reduce the memory and compute costs of large language model inference.
arXiv:2602. 01027v2 Announce Type: replace Abstract: Mixed-precision quantization is a promising approach for compressing large language models under tight memory budgets.
arXiv:2605. 26092v4 Announce Type: replace-cross Abstract: The deployment of Large Language Models (LLMs) and Vision Transformers (ViTs) on edge devices is significantly constrained by memory limitations and the critical timing bottlenecks introduced by dense Multiply-Accumulate (MAC) arrays.
arXiv:2410. 13056v4 Announce Type: replace-cross Abstract: Large Language Models (LLMs) have demonstrated remarkable success across a wide range of language tasks, but their deployment on edge devices remains challenging due to the substantial memory requirements imposed by their large parameter sizes.
arXiv:2608. 15602v1 Announce Type: cross Abstract: While binary quantization theoretically promises extreme compression and acceleration for Large Language Models (LLMs), existing research often overlooks the necessity of specialized hardware kernels, thus failing to unleash the full acceleration potential due to persistent reliance on expensive floating-point arithmetic or runtime dequantization overheads.
arXiv:2608. 06763v1 Announce Type: new Abstract: Weight quantization for large-language-model inference must balance adaptive reconstruction levels with representations regular enough for efficient GPU execution.
arXiv:2606. 15652v1 Announce Type: new Abstract: 4-bit quantization significantly reduces the memory footprint and accelerates the inference of large language models (LLMs).
Weight quantization for large-language-model inference must balance adaptive reconstruction levels with representations regular enough for efficient GPU execution. Uniform integers constrain each group to a linear grid.