Code LLMs correctly label incorrect repair instructions as wrong, then follow them anyway, creating compounding Ghost Errors that self-guided iterative repair usually cannot reverse.
Automating the correctness assessment of AI-generated code for security contexts , volume=
1 Pith paper cite this work, alongside 29 external citations. Polarity classification is still indexing.
1
Pith paper citing it
29
external citations · external index
fields
cs.SE 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Obey, Diverge, Collapse: Blind Obedience to Incorrect Instructions Drives Code LLMs to Irrecoverable Code Semantic Collapse
Code LLMs correctly label incorrect repair instructions as wrong, then follow them anyway, creating compounding Ghost Errors that self-guided iterative repair usually cannot reverse.