@misc{indiciae04f4a6484916,
  title = {Mitigating Over-Suppression in Speech Enhancement via Inference-Time Rethink-and-Refine Correction Module},
  author = {Mike Qu and Yu-Wen Chen and Julia Hirschberg},
  year = {2026},
  url = {https://arxiv.org/abs/2608.07781},
  note = {Source identifier: 2608.07781}
}