CAFE: Confining Feature Distillation to Language-Routed Adapters for Compact Multilingual Automatic Speech Recognition
arXiv:2610.19134v1 Announce Type: new Abstract: Feature distillation is usually expected to add to logit distillation. We find that this expectation does not hold for two common feature losses in compact multilingual automatic speech recognition (ASR), and analyze why through their gradients. The g…
Read the full story at arXiv cs.CL ↗
Timeline · 1 report
- 2026-10-15 04:00 · arXiv cs.CL
CAFE: Confining Feature Distillation to Language-Routed Adapters for Compact Multilingual Automatic Speech Recognition