Preprint
Aug 2026
Likelihood-Constrained Acoustic Reranking for Training-Free Hallucination Mitigation in LLM-Based ASR
Likelihood-Constrained Acoustic Reranking (LCAR), a training-free decoding method that improves acoustic grounding while preserving support from the base model, is proposed and evaluated on four LLM-based ASR systems using human-audited TTS and open-source speech challenge suites.
Ji-Shen Kuang, Lin-Ru Zheng, Hong-Jin Song et al.
· 0 citations