Attention-Guided Reliability Scaling for Contrastive Decoding in Robust Audio-Visual Speech Recognition

arXiv:2608.26213v1 Announce Type: cross Abstract: Large language model (LLM)-based audio-visual speech recognition (AVSR) systems are robust under noise. Contrastive decoding (CD), originally introduced to stabilize LLM generation by contrasting a weaker model against a stronger one at inference time, adjusts predictions without additional training. In this work, we apply CD to AVSR by contrasting audio-only conditioning with full audio-visual conditioning within the same underlying model. However, using a fixed contrastive strength introduces a trade-off across noise levels: stronger intervention helps under severe noise but may over-correct reliable predictions in clean conditions. We propose reliability-aware scaling of CD for AVSR. Instead of using a fixed strength, we adaptively modulate the contrastive influence at each token based on reliability signals derived from attention dynamics and inter-model predictive divergence. Experiments on LRS3 show consistent improvements across clean and low-SNR conditions.

Related

• VIB-AVSR: Variational Information Bottleneck for Noise-Robust LLM-Based Audio-Visual Speech Recognition
• Don't Let the Video Speak: Audio-Contrastive Preference Optimization for Audio-Visual Language Models
• Dr. SHAP-AV: Decoding Relative Modality Contributions via Shapley Attribution in Audio-Visual Speech Recognition
• Reliability-Guided Adaptive Ensembling for Robust Test-Time Adaptation

Source: arXiv cs.CV | 2026-08-28