LLMs show systematic output-mode collapse on closed-form prompts, with only ~22% of semantically equivalent variants preserving the requested bare-label format across five models and four tasks.
Automatically Generated Multi-Agent Framework for Jailbreaking Large Language Models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Paraphrase-Induced Output-Mode Collapse: When LLMs Break Character Under Semantically Equivalent Inputs
LLMs show systematic output-mode collapse on closed-form prompts, with only ~22% of semantically equivalent variants preserving the requested bare-label format across five models and four tasks.