paper-with-me

Papers

BitNet a4.8: 4-bit Activations for 1-bit LLMs

2024-11-07 · Hongyu Wang, Shuming Ma, Furu Wei

Recent research on the 1-bit Large Language Models (LLMs), such as BitNet b1.58, presents a promising direction for reducing the inference cost of LLMs while maintaining their performance. In this work, we introduce BitNet a4.8, enabling 4-bit activations for 1-bit LLMs. BitNet a4.8 employs a hybrid quantization and sparsification strategy to mitigate the quantization errors introduced by the outlier channels. Specifically, we utilize 4-bit activations for inputs to the attention and feed-forward network layers, while sparsifying intermediate states followed with 8-bit quantization. Extensive experiments demonstrate that BitNet a4.8 achieves performance comparable to BitNet b1.58 with equivalent training costs, while being faster in inference with enabling 4-bit (INT4/FP4) kernels. Additionally, BitNet a4.8 activates only 55% of parameters and supports 3-bit KV cache, further enhancing the efficiency of large-scale LLM deployment and inference.

📄 PDF Abstract BibTeX arXiv:2411.04965

Code (1)

microsoft/bitblas pytorch

Tasks

Quantization

Methods 이 논문이 사용한 방법론

Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention 설명 없음

Similar Papers 제목 키워드 기반

Q-Sparse: All Large Language Models can be Fully Sparsely-Activated

2024-07-15 · Hongyu Wang, Shuming Ma, Ruiping Wang, Furu Wei

We introduce, Q-Sparse, a simple yet effective approach to training sparsely-activated large language models (LLMs). Q-Sparse enables full sparsity of activations in LLMs which can bring significant efficiency gains in i…

All

1-bit AI Infra: Part 1.1, Fast and Lossless BitNet b1.58 Inference on CPUs

2024-10-21 · Jinheng Wang, Hansong Zhou, Ting Song, Shaoguang Mao 외

Recent advances in 1-bit Large Language Models (LLMs), such as BitNet and BitNet b1.58, present a promising approach to enhancing the efficiency of LLMs in terms of speed and energy consumption. These developments also e…

Bitnet.cpp: Efficient Edge Inference for Ternary LLMs

2025-02-17 · Jinheng Wang, Hansong Zhou, Ting Song, Shijie Cao 외

The advent of 1-bit large language models (LLMs), led by BitNet b1.58, has spurred interest in ternary LLMs. Despite this, research and practical applications focusing on efficient edge inference for ternary LLMs remain …

Sparse-BitNet: 1.58-bit LLMs are Naturally Friendly to Semi-Structured Sparsity

2026-03-05 · Di Zhang, Xun Wu, Shaohan Huang, Yudong Wang 외 arxiv

Semi-structured N:M sparsity and low-bit quantization (e.g., 1.58-bit BitNet) are two promising approaches for improving the efficiency of large language models (LLMs), yet they have largely been studied in isolation. In…

BitNet Distillation

2025-10-15 · Xun Wu, Shaohan Huang, Wenhui Wang, Ting Song 외 arxiv

In this paper, we present BitNet Distillation (BitDistill), a lightweight pipeline that fine-tunes off-the-shelf full-precision LLMs (e.g., Qwen) into 1.58-bit precision (i.e., ternary weights {-1, 0, 1}) for specific do…