Search papers, labs, and topics across Lattice.
King's College London, The Alan Turing Institute
3
0
5
LLM agents in decentralized marketplaces are demonstrably exploitable, with adversarial prompting enabling over 50% fraudulent transaction completion and doubling agent profits.
High IoU scores often conceal broken CAD construction logic鈥攁 blind spot resolved here by translating natural language into checkable constraint trees that can verify and self-repair parametric CAD code.
