SOURCE-LINKED INTELLIGENCE

REVE: Efficient Hallucination Correction for Large Audio-Language Models via Reused Encoder States

arXiv · AI, language, vision and robotics · article · Sep 22, 2026 · UTC

Large audio-language models may mention acoustic events that are absent from the input. A separate audio event detector can verify these mentions, but doing so requires a second audio encoder and a separate forward pass. We propose Reused Encoder States for Verifying Events (REVE), a lightweight method that uses states already computed by the target model. One readout summarizes class scores across audio frames, while another uses pooled states from four consecutive frame intervals. Class-aware score fusion combines their outputs to verify generated event mentions without encoding the audio ag

Read original source ↗ Open in workspace

recordType

paper

region

Global