6 papers
Agentic Data Sampling Framework for Multilingual ASR via Linguistic-Aware Error-Driven Optimization
Zhiheng Qian, Junhao Du, Jing Peng +6
Multilingual automatic speech recognition (ASR) has achieved rapid progress with large-scale pretrained models and multilingual speech corpora. Previous studies have proved the piv…
ETC-Bench: A Multilingual Benchmark for Evaluation-Side Transcript Canonicalization
Zhiheng Qian, Junhao Du, Jing Peng +5
Written transcripts of the same speech can differ in ways that are not uniquely determined by the speech signal. We formulate evaluation-side transcript canonicalization (ETC), whi…
A frugal primal-dual splitting with minimal lifting over arbitrary rooted trees
Feng Xue, Hui Zhang
We develop a frugal primal-dual splitting with minimal lifting for solving structured monotone inclusions, involving cocoercive operators, linear compositions and parallel sums. Th…
Dual-LoRA: Parameter-Efficient Adversarial Disentanglement for Cross-Lingual Speaker Verification
Qituan Shangguan, Junhao Du, Kunyang Peng +5
Cross-lingual speaker verification suffers from severe language-speaker entanglement. This causes systematic degradation in the hardest scenario: correctly accepting utterances fro…
Subspace decomposition in regularized least-squares: solution properties, restricted coercivity and beyond
Feng Xue, Hui Zhang
We investigate the solution properties of the regularized least-squares problem. Using a subspace decomposition technique, we derive expressions for the solution set in terms of th…
On degenerate preconditioned proximal point methods under restricted monotonicity
Feng Xue, Hui Zhang
This work investigates the fundamental properties of the degenerate preconditioned resolvent under restricted monotonicity. We extend key notions of non-expansiveness and demiclose…