2 papers
cs.CL2026
Writing-System-Level Tokenizer Adaptation for Byte-Level BPE
Bohdan Didenko
Pretrained byte-level BPE tokenizers can segment underrepresented languages inefficiently. Replacing a tokenizer changes the meaning of nearly every token ID, while vocabulary expa…
cs.CL2023
RedPenNet for Grammatical Error Correction: Outputs to Tokens, Attentions to Spans
Bohdan Didenko, Andrii Sameliuk
The text editing tasks, including sentence fusion, sentence splitting and rephrasing, text simplification, and Grammatical Error Correction (GEC), share a common trait of dealing w…