Search papers, labs, and topics across Lattice.
Vulcan Research, AIFT
5
0
8
3
AMT-X reveals that existing safety evaluations can miss up to 33% of actionable harm by conflating partial success with full operational capability.
Incentivizing honest participation in federated learning is now possible without ground truth labels, even when some participants are trying to game the system.
Existing security tools are blind to critical threat vectors in agentic systems, but MCPThreatHive offers an automated, open-source solution to illuminate and classify these risks.
Blockchain-based federated learning can be made practical by using multi-task peer prediction to overcome the computational bottleneck of contribution measurement.
Existing threat models fail to capture the unique vulnerabilities of Model Context Protocol systems, but MCP-38 fills this gap with a comprehensive taxonomy of 38 distinct threat categories.
