Search papers, labs, and topics across Lattice.
7
0
7
5
LLMs excel at generating functional RTL code but fail to meet security standards, with a mere 14-35% passing rate in security tests.
EAVA not only predicts software vulnerabilities but also provides actionable evidence, bridging the gap between automated assessments and human validation.
Execution signals and reasoning signals have orthogonal errors, and their independent calibration can significantly enhance Verilog code generation accuracy.
Refploit recovers 80.2% of Java vulnerability exploits by transforming failed agent trajectories into actionable insights, revealing the untapped potential of incomplete exploit attempts.
Turns out, users are more worried about LLM access and deployment than just generation quality.
Discovering exploitable library vulnerabilities in client code doesn't require proof-of-concept exploits anymore, thanks to a new fuzzing technique.
Automated patch porting can now handle twice as many inconsistencies by leveraging LLMs to reason about global codebase differences, not just local context.
