A beam-search-based testing method (ABS) generates adversarial text perturbations that flip LLM classifier labels more often and more cheaply than five existing attack methods across three datasets.
Chatgpt incorrectness detection in software reviews,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.SE 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
unclear 1representative citing papers
citing papers explorer
-
Assessing the Robustness of LLM-based NLP Software via Automated Testing
A beam-search-based testing method (ABS) generates adversarial text perturbations that flip LLM classifier labels more often and more cheaply than five existing attack methods across three datasets.