BrailleLLM: Braille Instruction Tuning with Large Language Models for Braille Domain Tasks
This work addresses Braille domain tasks for visually impaired individuals, but it is incremental as it builds on existing large language models with tailored fine-tuning and datasets.
The paper tackled the problem of Braille information processing, which suffers from data scarcity and ambiguities in mixed-text contexts, by constructing English and Chinese Braille Mixed Datasets with mathematical formulas and proposing a syntax tree-based augmentation method, resulting in BrailleLLM achieving significant performance improvements in Braille translation scenarios over conventional fine-tuning.
Braille plays a vital role in education and information accessibility for visually impaired individuals. However, Braille information processing faces challenges such as data scarcity and ambiguities in mixed-text contexts. We construct English and Chinese Braille Mixed Datasets (EBMD/CBMD) with mathematical formulas to support diverse Braille domain research, and propose a syntax tree-based augmentation method tailored for Braille data. To address the underperformance of traditional fine-tuning methods in Braille-related tasks, we investigate Braille Knowledge-Based Fine-Tuning (BKFT), which reduces the learning difficulty of Braille contextual features. BrailleLLM employs BKFT via instruction tuning to achieve unified Braille translation, formula-to-Braille conversion, and mixed-text translation. Experiments demonstrate that BKFT achieves significant performance improvements over conventional fine-tuning in Braille translation scenarios. Our open-sourced datasets and methodologies establish a foundation for low-resource multilingual Braille research.