Embedding Layer Surgery and Task-wise Beam Search reduce catastrophic forgetting when adding new languages to Whisper, lowering old-language AWER from 14.2% to 11.9% versus Experience Replay.
Language-agnostic multilingual modeling,
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
background 1representative citing papers
citing papers explorer
-
Continual Learning with Embedding Layer Surgery and Task-wise Beam Search using Whisper
Embedding Layer Surgery and Task-wise Beam Search reduce catastrophic forgetting when adding new languages to Whisper, lowering old-language AWER from 14.2% to 11.9% versus Experience Replay.