LingoLoop traps MLLMs into generating up to 367 times more tokens by applying POS-aware attention adjustments to postpone EOS tokens and pruning generative paths to sustain repetitive loops.
An engorgio prompt makes large language model babble on
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
representative citing papers
Poisoning external knowledge bases with LLM-agent-crafted documents can increase RAG inference token consumption by up to 13.12 times at over 90% success rate while preserving answer quality.
citing papers explorer
-
LingoLoop Attack: Trapping MLLMs via Linguistic Context and State Entrapment into Endless Loops
LingoLoop traps MLLMs into generating up to 367 times more tokens by applying POS-aware attention adjustments to postpone EOS tokens and pruning generative paths to sustain repetitive loops.
-
Inference Cost Attacks for Retrieval-Augmented Large Language Models
Poisoning external knowledge bases with LLM-agent-crafted documents can increase RAG inference token consumption by up to 13.12 times at over 90% success rate while preserving answer quality.