Pith. sign in

Title resolution pending

1 Pith paper cite this work. Polarity classification is still indexing.

1 Pith paper citing it

fields

cs.AI 1

years

2025 1

verdicts

UNVERDICTED 1

representative citing papers

PUZZLED: Jailbreaking LLMs through Word-Based Puzzles

cs.AI · 2025-08-02 · unverdicted · novelty 6.0

PUZZLED masks harmful keywords as word-search, anagram, or crossword puzzles and achieves a reported 88.8% average attack success rate across five leading LLMs.

citing papers explorer

Showing 1 of 1 citing paper.

  • PUZZLED: Jailbreaking LLMs through Word-Based Puzzles cs.AI · 2025-08-02 · unverdicted · none · ref 7

    PUZZLED masks harmful keywords as word-search, anagram, or crossword puzzles and achieves a reported 88.8% average attack success rate across five leading LLMs.