VARAN dynamically reweights internal layer representations per input through probing heads and data-dependent weights, reporting improved ASR and emotion recognition with LoRA.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
VARAN: Variational Inference for Self-Supervised Speech Models Fine-Tuning on Downstream Tasks
VARAN dynamically reweights internal layer representations per input through probing heads and data-dependent weights, reporting improved ASR and emotion recognition with LoRA.