CASCADE finds code-documentation mismatches by running LLM-generated tests from docs and confirming failure only when documentation-derived code succeeds on the same test.
Assess- ing Hidden Risks of LLMs: An Empirical Study on Robustness, Consistency, and Credibility.CoRR abs/2305.10235
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2verdicts
UNVERDICTED 2representative citing papers
PopQuiz Attack infers LLM training data membership by turning examples into quiz questions and measuring answer accuracy, reaching 0.873 average ROC-AUC across six models and outperforming prior methods by 20.6%.
citing papers explorer
-
CASCADE: Detecting Inconsistencies between Code and Documentation with Automatic Test Generation
CASCADE finds code-documentation mismatches by running LLM-generated tests from docs and confirming failure only when documentation-derived code succeeds on the same test.
-
Pop Quiz Attack: Black-box Membership Inference Attacks Against Large Language Models
PopQuiz Attack infers LLM training data membership by turning examples into quiz questions and measuring answer accuracy, reaching 0.873 average ROC-AUC across six models and outperforming prior methods by 20.6%.