Tag
论文提出MixLID，一个支持111种语言-文字组合的token级混合语言识别模型，采用两阶段训练（先单语再混合语言数据），在单语和混合语言识别基准上均超越现有基线，模型已在Hugging Face开源。
This paper argues that the loanword-vs-switch annotation boundary, rather than the choice of model, drives Kazakh-Russian code-switching identification. The authors present a document-level gold LID dataset with an explicit annotation rule and show that naive heuristics and off-the-shelf LID tools fail to distinguish integrated borrowings from genuine code-switches.
This paper addresses open questions in the Gold-Angluin model of language identification in the limit, showing that computational traces using only a small alphabet and defined directly from the language enable identification in the limit, without requiring an underlying machine model.
This paper proposes two hybrid continual learning methods—Replay Augmented Elastic Weight Consolidation and Constraint Guided Knowledge Distillation—to adapt pretrained speech models for identifying low-resource Australian Aboriginal languages while mitigating catastrophic forgetting.
This study evaluates bilingual fine-tuning with language identification tokens for improving ASR in low-resource languages across nine diverse language pairs, finding that high LID accuracy is beneficial and that providing the LID token at inference can boost performance when LID accuracy is low.
GitHub announces the GitHub Multilingual Repositories Dataset, an open metadata dataset covering over 80 million classification rows across 40 million repositories to help researchers and developers build multilingual AI tools.