Re-encoding UTF-8 byte fallbacks as shared 6-bit prefixes plus 9-bit tokens shortens CJK token sequences losslessly, at the cost of reduced tokenizer entropy and mixed wall-clock speedups.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Bit-level BPE: Below the byte boundary
Re-encoding UTF-8 byte fallbacks as shared 6-bit prefixes plus 9-bit tokens shortens CJK token sequences losslessly, at the cost of reduced tokenizer entropy and mixed wall-clock speedups.