SOURCE-LINKED INTELLIGENCE

Reducing Hallucinated Transcripts in Whisper via Hallucination Space Projection

arXiv · AI, language, vision and robotics · article · Sep 3, 2026 · UTC

Whisper is a widely used foundation model for automatic speech recognition (ASR), but its generative decoder can produce fluent hallucinated transcripts for inputs containing little or no speech. We propose a training-free, inference-time method to reduce these hallucinations using low-rank projection of decoder activations. A compact hallucination-associated subspace is estimated from non-speech calibration data, and decoder hidden states are projected away from this subspace during inference. We evaluate two variants: always-on, which applies projection to all inputs, and gated, which applie

Read original source ↗ Open in workspace

recordType

paper

region

Global