Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Affiliation:
3
0
7
9
Despite the promise of LLMs, they only recover a fraction of indicators in CTI generation, highlighting the need for expert oversight in automated intelligence production.
LLMs struggle significantly at comprehension stages in reverse engineering, revealing a competence limit that challenges assumptions about model size and cost.
LLMs are not just tools; they are susceptible to cognitive biases that can drastically undermine their effectiveness in detecting code vulnerabilities.