We have not lifted any functions out of this paper's repositories yet, so there is nothing we have run. The repositories linked to it are listed below.
Some links come from the archived Papers with Code dataset (CC BY-SA 4.0): attribution and licence.
Jailbreaking techniques trick Large Language Models (LLMs) into producing restricted output, posing a potential threat. One line of defense is to use another LLM as a Judge to evaluate the harmfulness of generated text. However, we reveal that these Judge LLMs are vulnerable to token segmentation bias, an issue that arises when delimiters alter the tokenization process, splitting words into smaller sub-tokens. This alters the embeddings of the entire sequence, reducing detection accuracy and allowing harmful content to be misclassified as safe. In this paper, we introduce Emoji Attack, a novel strategy that amplifies existing jailbreak prompts by exploiting token segmentation bias. Our method leverages in-context learning to systematically insert emojis into text before it is evaluated by a Judge LLM, inducing embedding distortions that significantly lower the likelihood of detecting unsafe content. Unlike traditional delimiters, emojis also introduce semantic ambiguity, making them particularly effective in this attack. Through experiments on state-of-the-art Judge LLMs, we demonstrate that Emoji Attack substantially reduces the unsafe prediction rate, bypassing existing safeguards.
The same record, over MCP at https://syntology.ai/mcp:
get_harvested_code_for_paper("2411.01077")
get_code_for_paper("2411.01077")
have("2411.01077")
The run record, dated, one paper per request, free:
curl https://syntology.ai/api/ran/2411.01077.json
A badge for a README (the split and the date, never a ratio):
[](https://syntology.ai/paper/2411.01077)
Connect an agent — have() is free.