重新思考脉冲Transformer中的成对Token交互
Rethinking Pairwise Token Interaction in Spiking Transformers
浏览论文内容
中文总结 AI 辅助
针对脉冲Transformer中成对token交互的稀疏性问题，提出GSAP机制，通过轴向传播与接收器条件门控解耦信息传播与上下文选择，实现结构化长距离通信，同时保持脉冲表示的稀疏事件驱动特性。
中文摘要 AI 辅助
脉冲Transformer继承了传统Transformer的token交互机制，但其稀疏的二元表示从根本上改变了token间通信的建立方式。特别是，基于脉冲的查询-键匹配产生了高度稀疏且依赖于输入的交互模式，将信息传播与匹配脉冲事件的即时可用性耦合在一起。这促使了一种不同的交互范式，其中长距离通信不仅仅依赖于成对脉冲重合。因此，我们提出了门控脉冲轴向传播（GSAP），一种脉冲原生的token交互机制，将信息传播与上下文选择解耦。GSAP不通过查询-键匹配直接确定通信，而是首先沿水平和垂直轴传播基于脉冲的上下文，使信息通过结构化的顺序传播到达远处的token。然后，一个接收器条件门控决定在每个token处合并多少传播的上下文，而一个轻量级的局部路径则保留细粒度的邻域信息。通过这种方式，GSAP将token交互重新表述为先传播后选择的过程，在保持脉冲表示的稀疏事件驱动特性的同时，实现了结构化的长距离通信。代码可在以下网址获取：此https URL。
英文摘要
Spiking Transformers inherit token interaction mechanisms from conventional Transformers, yet their sparse binary representations fundamentally alter how token-to-token communication is established. In particular, spike-based query-key matching produces highly sparse and input-dependent interaction patterns, coupling information propagation to the instantaneous availability of matching spike events. This motivates a different interaction paradigm in which long-range communication does not rely solely on pairwise spike coincidence. We therefore propose Gated Spike Axial Propagation (GSAP), a spike-native token interaction mechanism that decouples information propagation from context selection. Instead of directly determining communication through query-key matching, GSAP first propagates spike-based context along the horizontal and vertical axes, allowing information to reach distant tokens through structured sequential propagation. A receiver-conditioned gate then determines how much of the propagated context is incorporated at each token, while a lightweight local pathway preserves fine-grained neighborhood information. In this way, GSAP reformulates token interaction as a propagate-then-select process, enabling structured long-range communication while retaining the sparse event-driven nature of spiking representations. Code is available at https://github.com/Fancyssc/GSAP.
发表机构
- Institute of Automatoin, CAS（中国科学院自动化研究所）
- Zhongguancun Academy（中关村学院）
机构由 AI 辅助整理，请以论文原文为准。
