Search papers, labs, and topics across Lattice.
1
0
3
LLMs can achieve near-perfect F1 scores (>=0.978) in detecting interprocedural C vulnerabilities for under $0.60, but the best model varies by language.