端到端口语语言理解中的选择性能力遗忘
Selective Capability Unlearning in End-to-End Spoken Language Understanding
AI总结：
提出绑定子空间（BSU）框架，通过隔离和衰减意图条件方向，解决自回归模型中意图抑制后条件映射残留的能力持久性问题，在SLU基准上显著降低强制前缀可恢复性。
摘要：
现代口语语言理解（SLU）系统越来越多地部署在现实环境中，由于策略或安全约束，可能需要移除特定功能。在SLU中，一个功能对应一个意图及其相关的槽生成行为。然而，在自回归模型中，抑制目标意图并不会消除以该意图为条件的槽生成条件映射。当意图前缀由外部提供时，模型可以重建原始的意图-槽结构。我们将这种结构性失败称为\textbf{\emph{能力持久性}}。我们提出\textit{\underline{绑定子空间（BSU）}}，一种表示级框架，用于隔离和衰减该映射背后的意图条件方向。在SLU基准测试中，BSU显著降低了强制前缀可恢复性，同时保持了保留性能。
英文摘要：
Modern spoken language understanding (SLU) systems are increasingly deployed in real-world settings, where specific functionalities may need to be removed due to policy or safety constraints. In SLU, a functionality corresponds to an intent and its associated slot-generation behavior. However, in autoregressive models, suppressing a target intent does not eliminate the conditional mapping that generates slots conditioned on that intent. When the intent prefix is externally supplied, the model can reconstruct the original intent-slot structure. We identify this structural failure as \textbf{\emph{capability persistence}}. We propose \textit{\underline{B}inding \underline{S}ubspace (BSU)}, a representation-level framework that isolates and attenuates intent-conditioned directions underlying this mapping. Across SLU benchmarks, BSU substantially reduces forced-prefix recoverability while preserving retained performance.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
