paper-with-me

홈 › Papers

What and When to Learn: CURriculum Ranking Loss for Large-Scale Speaker Verification

2026-03-25 · Massa Baali, Sarthak Bisht, Rita Singh, Bhiksha Raj arxiv

Speaker verification at large scale remains an open challenge as fixed-margin losses treat all samples equally regardless of quality. We hypothesize that mislabeled or degraded samples introduce noisy gradients that disrupt compact speaker manifolds. We propose Curry (CURriculum Ranking), an adaptive loss that estimates sample difficulty online via Sub-center ArcFace: confidence scores from dominant sub-center cosine similarity rank samples into easy, medium, and hard tiers using running batch statistics, without auxiliary annotations. Learnable weights guide the model from stable identity foundations through manifold refinement to boundary sharpening. To our knowledge, this is the largest-scale speaker verification system trained to date. Evaluated on VoxCeleb1-O, and SITW, Curry reduces EER by 86.8\% and 60.0\% over the Sub-center ArcFace baseline, establishing a new paradigm for robust speaker verification on imperfect large-scale data.

📄 PDF Abstract BibTeX arXiv:2603.24432

Code (0)

등록된 구현이 없습니다.

Tasks

Speaker Verification

Similar Papers 제목 키워드 기반

Training Curricula for Open Domain Answer Re-Ranking

2020-04-29 · Sean MacAvaney, Franco Maria Nardini, Raffaele Perego, Nicola Tonellotto 외

In precision-oriented tasks like answer ranking, it is more important to rank many relevant answers highly than to retrieve all relevant answers. It follows that a good ranking strategy would be to learn how to identify …

Re-Ranking

Curriculum Discovery through an Encompassing Curriculum Learning Framework

2021-09-29 · Mohamed Elgaar, Hadi Amiri

We describe a curriculum learning framework capable of discovering optimal curricula in addition to performing standard curriculum learning. We show that this framework encompasses existing curriculum learning approaches…

Natural Language Inferencetext-classificationText Classification

Mancs: A Multi-task Attentional Network with Curriculum Sampling for Person Re-identification

2018-09-01 · ECCV 2018 9 · Cheng Wang, Qian Zhang, Chang Huang, Wenyu Liu 외

We propose a novel deep network called Mancs that solves the person re-identification problem from the following aspects: fully utilizing the attention mechanism for the person misalignment problem and properly sampling …

Person Re-Identification

What do Language Models Learn and When? The Implicit Curriculum Hypothesis

2026-04-09 · Emmy Liu, Kaiser Sun, Millicent Li, Isabelle Lee 외 arxiv

Large language models (LLMs) can perform remarkably complex tasks, yet the fine-grained details of how these capabilities emerge during pretraining remain poorly understood. Scaling laws on validation loss tell us how mu…

Logical Reasoning

When to Learn What: Model-Adaptive Data Augmentation Curriculum

2023-09-09 · ICCV 2023 1 · Chengkai Hou, Jieyu Zhang, Tianyi Zhou

Data augmentation (DA) is widely used to improve the generalization of neural networks by enforcing the invariances and symmetries to pre-defined transformations applied to input data. However, a fixed augmentation polic…

Data AugmentationFairnessimage-classificationImage Classification