1 paper
Yanjie Zhao, Hongjie Chen, Li Lu +3
Ask a large language model (LLM) whether a pointer dereference is safe, and it can often produce a plausible justification for ``yes''. The difficulty is that a fluent justificatio…