# TiSpell: A Semi-Masked Methodology for Tibetan Spelling Correction covering Multi-Level Error with Data Augmentation

arXiv: https://arxiv.org/abs/2505.08037 · HTML: https://mlsanity.com/paper/2505.08037/

## Key facts

- Cited by 2 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 43 references that are in the map.
- First posted on arXiv 12 May 2025.

## Builds on (most cited first, 18 of 18)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Learning Phrase Representations using RNN Encoder-Decoder for Statistical Machine Translation](https://mlsanity.com/paper/1406.1078/) (2014)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [Deep contextualized word representations](https://mlsanity.com/paper/1802.05365/) (2018)
- [SentencePiece: A simple and language independent subword tokenizer and detokenizer for Neural Text Processing](https://mlsanity.com/paper/1808.06226/) (2018)
- [Google's Neural Machine Translation System: Bridging the Gap between Human and Machine Translation](https://mlsanity.com/paper/1609.08144/) (2016)
- [Evaluation: from precision, recall and F-measure to ROC, informedness, markedness and correlation](https://mlsanity.com/paper/2010.16061/) (2020)
- [Character-Aware Neural Language Models](https://mlsanity.com/paper/1508.06615/) (2015)
- [Spelling Error Correction with Soft-Masked BERT](https://mlsanity.com/paper/2005.07421/) (2020)
- [Encoder-Decoder Models Can Benefit from Pre-trained Masked Language Models in Grammatical Error Correction](https://mlsanity.com/paper/2005.00987/) (2020)
- [NeuSpell: A Neural Spelling Correction Toolkit](https://mlsanity.com/paper/2010.11085/) (2020)
- [Rethinking Masked Language Modeling for Chinese Spelling Correction](https://mlsanity.com/paper/2305.17721/) (2023)
- [Robsut Wrod Reocginiton via semi-Character Recurrent Neural Network](https://mlsanity.com/paper/1608.02214/) (2016)
- [TURNA: A Turkish Encoder-Decoder Language Model for Enhanced Understanding and Generation](https://mlsanity.com/paper/2401.14373/) (2024)
- [AraSpell: A Deep Learning Approach for Arabic Spelling Correction](https://mlsanity.com/paper/2405.06981/) (2024)
- [Spelling Error Correction Using a Nested RNN Model and Pseudo Training Data](https://mlsanity.com/paper/1811.00238/) (2018)
- [DPCSpell: A Transformer-based Detector-Purificator-Corrector Framework for Spelling Error Correction of Bangla and Resource Scarce Indic Languages](https://mlsanity.com/paper/2211.03730/) (2022)

## Built on by (most cited first, 2 of 2)

- [TIBSTC-CoT: A Multi-Domain Instruction Dataset for Chain-of-Thought Reasoning in Language Models](https://mlsanity.com/paper/2508.01977/) (2025)
- [Tibetan Language and AI: A Comprehensive Survey of Resources, Methods and Challenges](https://mlsanity.com/paper/2510.19144/) (2025)

## Abstract (arXiv)

Multi-level Tibetan spelling correction addresses errors at both the character and syllable levels within a unified model. Existing methods focus mainly on single-level correction and lack effective integration of both levels. Moreover, there are no open-source datasets or augmentation methods tailored for this task in Tibetan. To tackle this, we propose a data augmentation approach using unlabeled text to generate multi-level corruptions, and introduce TiSpell, a semi-masked model capable of correcting both character- and syllable-level errors. Although syllable-level correction is more challenging due to its reliance on global context, our semi-masked strategy simplifies this process. We synthesize nine types of corruptions on clean sentences to create a robust training set. Experiments on both simulated and real-world data demonstrate that TiSpell, trained on our dataset, outperforms baseline models and matches the performance of state-of-the-art approaches, confirming its effectiveness.