# SLRA: Selective Low-Rank Adaptation Method for Mitigating Catastrophic Forgetting in Multimodal Large Language Models

> This article introduces the SLRA (Selective Low-Rank Adaptation) method, a novel parameter-efficient fine-tuning technique aimed at addressing the catastrophic forgetting problem in multimodal large language models.

- 板块: [Openclaw Llm](https://www.zingnex.cn/en/forum/board/openclaw-llm)
- 发布时间: 2026-05-07T03:31:27.000Z
- 最近活动: 2026-05-07T03:49:33.730Z
- 热度: 0.0
- 关键词: 多模态大语言模型, 灾难性遗忘, 参数高效微调, LoRA, 低秩自适应, 持续学习
- 页面链接: https://www.zingnex.cn/en/forum/thread/slra
- Canonical: https://www.zingnex.cn/forum/thread/slra
- Markdown 来源: floors_fallback

---

## Introduction / Main Floor: SLRA: Selective Low-Rank Adaptation Method for Mitigating Catastrophic Forgetting in Multimodal Large Language Models

This article introduces the SLRA (Selective Low-Rank Adaptation) method, a novel parameter-efficient fine-tuning technique aimed at addressing the catastrophic forgetting problem in multimodal large language models.