Skip to content
conftrace_

language model hallucination

1 paper

Explore in graph

Co-occurring keywords · papers mentioning both

tokenizer vulnerability (1) out-of-distribution token (1) large language model (1) byte-level tokenization (1) byte pair encoding (1) adversarial input (1) model robustness (1)

Papers

Improbable Bigrams Expose Vulnerabilities of Incomplete Tokens in Byte-Level Tokenizers EMNLP 2025