Liger：将大型语言模型线性化为门控循环结构

Liger: Linearizing Large Language Models to Gated Recurrent Structures

March 3, 2025

作者: Disen Lan, Weigao Sun, Jiaxi Hu, Jusen Du, Yu Cheng

cs.AI

摘要

采用线性循环建模的Transformer架构提供了线性时间训练和恒定内存推理的优势。尽管这些非标准架构已展现出高效性和性能，但从头开始预训练此类模型仍然成本高昂且风险较大。大型语言模型（LLM）的线性化将预训练的标准模型转化为线性循环结构，从而实现了更高效的部署。然而，当前的线性化方法通常需要引入额外的特征映射模块，这些模块需要大量微调，并且忽视了在顶尖线性循环模型中使用的门控机制。为解决这些问题，本文提出了Liger（线性化LLM至门控循环结构的简称），这是一种将预训练LLM转换为门控线性循环模型的新方法，无需添加额外参数。Liger通过重新利用预训练的关键矩阵权重来构建多样化的门控机制，促进了多种门控循环结构的形成，同时避免了从头训练额外组件的需求。利用低秩适应（LoRA）进行轻量级微调，Liger恢复了线性化门控循环模型的性能，使其与原始LLM相当。此外，我们引入了Liger Attention，一种层内混合注意力机制，在仅使用0.02%预训练令牌的情况下，显著恢复了基于Transformer的LLM 93%的性能，在多个基准测试中取得了竞争性结果，这一成果在1B至8B参数规模的模型上得到了验证。代码已发布于https://github.com/OpenSparseLLMs/Linearization。

English

Transformers with linear recurrent modeling offer linear-time training and constant-memory inference. Despite their demonstrated efficiency and performance, pretraining such non-standard architectures from scratch remains costly and risky. The linearization of large language models (LLMs) transforms pretrained standard models into linear recurrent structures, enabling more efficient deployment. However, current linearization methods typically introduce additional feature map modules that require extensive fine-tuning and overlook the gating mechanisms used in state-of-the-art linear recurrent models. To address these issues, this paper presents Liger, short for Linearizing LLMs to gated recurrent structures. Liger is a novel approach for converting pretrained LLMs into gated linear recurrent models without adding extra parameters. It repurposes the pretrained key matrix weights to construct diverse gating mechanisms, facilitating the formation of various gated recurrent structures while avoiding the need to train additional components from scratch. Using lightweight fine-tuning with Low-Rank Adaptation (LoRA), Liger restores the performance of the linearized gated recurrent models to match that of the original LLMs. Additionally, we introduce Liger Attention, an intra-layer hybrid attention mechanism, which significantly recovers 93\% of the Transformer-based LLM at 0.02\% pre-training tokens during the linearization process, achieving competitive results across multiple benchmarks, as validated on models ranging from 1B to 8B parameters. Code is available at https://github.com/OpenSparseLLMs/Linearization.