Periodically averaging per-dataset LoRA adapters into a fixed Whisper base reduces catastrophic forgetting in rehearsal-free continual learning for code-switched ASR, with modest gains over a weight-averaging plus distillation baseline.
In the factoriza- tion phase, we aim to expand the knowledge base model using temporary, learnable adapters on the datasets to be learned
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Weight Factorization and Centralization for Continual Learning in Speech Recognition
Periodically averaging per-dataset LoRA adapters into a fixed Whisper base reduces catastrophic forgetting in rehearsal-free continual learning for code-switched ASR, with modest gains over a weight-averaging plus distillation baseline.