# BPE
4 articlestagged with “BPE”
Tokenizer Security
How tokenization creates attack surfaces in LLM systems: BPE exploitation, token boundary attacks, encoding edge cases, and tokenizer-aware adversarial techniques.
tokenizerBPEencodingtoken-boundaryattack-surface
Tokenizer Attack Surface Analysis
Deep analysis of tokenizer vulnerabilities including token boundary exploitation, special token manipulation, and cross-tokenizer attacks.
modelstokenizerBPEattacks
Tokenizer Manipulation & Custom Vocabularies
Attacking BPE training data to influence vocabulary construction, inserting special tokens, manipulating merge rules, and creating custom tokenizer backdoors.
tokenizerBPEvocabularymerge-rulestoken-manipulationspecial-tokens
Token Smuggling
Exploiting LLM tokenization quirks to smuggle harmful content past safety filters by manipulating how text is split into tokens at the subword level.
jailbreakingtokenizationtoken-smugglingBPEsubwordcontent-filter-bypassred-teaming