Using Chain-of-Thought and Self-Consistency prompts, some LLMs removed over half of SAST false positives on a benchmark while missing no genuine weaknesses, and ensembling three models removed about 79%.
Learning a classifier for false positive error reports emitted by static code analysis tools,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CR 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Towards Effective Complementary Security Analysis using Large Language Models
Using Chain-of-Thought and Self-Consistency prompts, some LLMs removed over half of SAST false positives on a benchmark while missing no genuine weaknesses, and ensembling three models removed about 79%.