Mechanistic Insights into Context Failures of Audio-Language Models for Impaired Speech
This work introduces Self-Contrastive Residual Alignment (SCRA), a one-pass inference-time edit that learns a low-rank linear corrector for the prompt-induced residual shift and slightly but significantly improves over the audio-only baseline on difficult samples.
P. Moure, Bilal Bounajma, Niclas Pokel et al.
· 0 citations