用原型向量替代注意力机制，让语言模型推理更可解释。
Prototype Transformer: Towards Language Model Architectures Interpretable by Design
- 用可学习的原型向量替代Transformer的二次复杂度注意力
- 原型自动捕捉'女性'等可命名概念，支持行为编辑
- 在文本生成和GLUE任务上表现优异，适合追求可解释性的研究者
当前顶尖语言模型虽在某些领域超越人类，但其推理过程仍不透明，降低可信度并增加欺骗与幻觉风险。我们提出原型Transformer（ProtoT），一种自回归语言模型架构，将Transformer中二次计算成本的自注意力模块替换为基于原型的线性成本模块，原型是可学习的参数向量。在ProtoT中，原型构建了不同时间尺度的信息聚合通道。实验表明，该结构使原型在训练过程中自动捕获如'woman'等可命名概念，为理解模型推理和针对性修改模型行为提供了可能。相比基线方法，ProtoT在模型规模和数据量增长时仍具良好的扩展性，对输入扰动鲁棒，并在文本生成及下游任务（包括GLUE）上表现良好。这些结果表明，ProtoT是迈向可设计可解释自回归语言模型的重要一步。
原文摘要 · Abstract (English)
While state-of-the-art language models (LMs) surpass most humans in certain domains, their reasoning remains largely opaque, reducing trust and increasing the risk of deception and hallucination. We introduce the Prototype Transformer (ProtoT), an autoregressive LM architecture that replaces the quadratic-cost self-attention module of the Transformer with a linear-cost module based on prototypes, which are learned parameter vectors. In ProtoT, prototypes create communication channels that aggregate contextual information at different time scales. We show that this structure leads prototypes to automatically capture nameable concepts, such as "woman", during training, offering a path toward interpreting model reasoning and making targeted edits to model behavior. Compared with baselines, ProtoT scales well with model and data size, is robust to input perturbations, and performs well on text generation and downstream tasks, including GLUE. These results suggest that ProtoT is a promising step toward autoregressive language models that are more interpretable by design.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。