Transformer^2：自适应语言模型

Transformer^2: Self-adaptive LLMs

January 9, 2025

作者: Qi Sun, Edoardo Cetin, Yujin Tang

cs.AI

摘要

自适应大型语言模型（LLMs）旨在解决传统微调方法所面临的挑战，这些方法通常在计算上很昂贵，并且在处理多样化任务时缺乏灵活性。我们引入了\implname，这是一个新颖的自适应框架，通过有选择地调整它们的权重矩阵的单个组件，实时地为未知任务调整LLMs。在推断过程中，\implname采用两步机制：首先，一个调度系统识别任务属性，然后使用强化学习训练的任务特定“专家”向量被动态混合，以获得针对输入提示的定向行为。我们的方法在参数更少、效率更高的情况下胜过了常见方法，如LoRA。\implname在不同的LLM架构和模态，包括视觉-语言任务中展现了多样性。\implname代表了一个重大进步，为增强LLMs的适应性和任务特定性能提供了一个可扩展、高效的解决方案，为真正动态、自组织的人工智能系统铺平了道路。

English

Self-adaptive large language models (LLMs) aim to solve the challenges posed by traditional fine-tuning methods, which are often computationally intensive and static in their ability to handle diverse tasks. We introduce \implname, a novel self-adaptation framework that adapts LLMs for unseen tasks in real-time by selectively adjusting only the singular components of their weight matrices. During inference, \implname employs a two-pass mechanism: first, a dispatch system identifies the task properties, and then task-specific "expert" vectors, trained using reinforcement learning, are dynamically mixed to obtain targeted behavior for the incoming prompt. Our method outperforms ubiquitous approaches such as LoRA, with fewer parameters and greater efficiency. \implname demonstrates versatility across different LLM architectures and modalities, including vision-language tasks. \implname represents a significant leap forward, offering a scalable, efficient solution for enhancing the adaptability and task-specific performance of LLMs, paving the way for truly dynamic, self-organizing AI systems.