Residual quantization with four hierarchical codebooks improves BEATs-style bidirectional self-supervised pretraining over vector quantization on audio, image, and video benchmarks.
BEATs : audio pre-training with acoustic tokenizers
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
BRIDLE: Generalized Self-supervised Learning with Quantization
Residual quantization with four hierarchical codebooks improves BEATs-style bidirectional self-supervised pretraining over vector quantization on audio, image, and video benchmarks.