Selective Capability Unlearning in End-to-End Spoken Language Understanding

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 $\mathit{\text{capability}}\mathit{\text{persistence}}$. We propose $\underset{‾}{\mathit{\text{B}}}\mathit{\text{inding}}\underset{‾}{\mathit{\text{S}}}\mathit{\text{ubspace}}\mathit{\text{(BSU)}}$, 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.