arXiv Computation and Language
Phonological Knowledge of Chinese Characters in Language Models
Read the original on arXiv Computation and Language →The Flow has not summarised this story yet — read it at arXiv Computation and Language.
Related stories
arXiv Computation and Language
Sep 21
Beyond Atomic Tokens: Factorizing Syllables for Language Model Pretraining
arXiv:2609. 21362v1 Announce Type: new Abstract: Conventional tokenizers represent text as characters or statistically derived subwords, overlooking the internal phonological structure of syllables and often requiring large vocabularies.
arXiv AI
Jun 17
Perceptual compensation for tonal context in self-supervised speech models
arXiv:2606. 17835v1 Announce Type: cross Abstract: This study examines the extent to which the wav2vec2.
Hugging Face Trending Papers
Jun 24
Phonetic and semantic analyses of spoken corpora of Beijing and Taiwan Mandarin indicate that the neutral tone is a lexical tone
The neutral, or floating, tone of Mandarin Chinese is a tone with an enigmatic set of properties. It has been described as a reduced tone, or as a tone that sometimes is lexically fixed but that can also be toneless.
arXiv Computation and Language
Oct 23
Direct cueing of word identity through word duration: evidence from Mandarin homophones
The study investigates whether word duration can directly cue word identity in Mandarin Chinese homophones. Participants used duration differences to order homophone pairs, successfully mapping shorter durations to higher-frequency words even under high noise or with no other cues. These findings suggest that word duration is a phonetic detail incorporated into lexical representations.
arXiv Machine Learning
Oct 9
Phonologically Informed Tokenization for German Speech Recognition: A Cross-Domain Study
The study evaluates phonologically informed tokenization for German speech recognition using a wav2vec 2.0 backbone fine‑tuned with CTC. Three tokenizer families—pretrained multilingual characters, data‑driven BPE, and phonologically informed units from syllabification and grapheme‑to‑phoneme conversion—are compared across 40 fine‑tunes on three German test sets (in‑domain read, dialectal spontaneous, and standard spontaneous speech). Results show that phonologically informed tokenizers match BPE and character baselines in‑domain, but under domain shift their performance depends on vocabulary size rather than linguistic variation, with syllable‑aware tokenization benefiting dialectal and spontaneous speech at small vocabularies; a phoneme‑level confusion analysis indicates that the acoustic encoder dominates error patterns.
arXiv Computation and Language
Oct 8
Tiny-Scale Chinese BERT Pretraining: A Controlled Comparison of MLM, WWM, and MacBERT Strategies
arXiv:2610.08879v1 Announce Type: new Abstract: Pretraining strategies significantly impact the quality of language models, yet existing comparisons of Masked Language Modeling (MLM), Whole Word Mask...