SpectralFT fine-tunes only the top singular vectors of pre-trained Transformer weights for speaker verification, achieving modest EER reductions over LoRA on two benchmarks.
Adaptive budget allocation for parameter-efficient fine-tuning,
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
eess.AS 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Spectral-Aware Low-Rank Adaptation for Speaker Verification
SpectralFT fine-tunes only the top singular vectors of pre-trained Transformer weights for speaker verification, achieving modest EER reductions over LoRA on two benchmarks.