Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Affiliation:
5
0
8
2
Clients can achieve superior worst-decile accuracy in federated learning by stopping local training based on an optimal stopping framework that balances loss reduction against drift, outperforming tuned baselines and existing adaptive methods.
Target-aligned input reparameterization slashes prediction errors for thermodynamic properties in supercritical combustion by up to 14.5 times, revolutionizing neural network efficiency.
By leveraging a unified temporal graph, FuseChain uncovers hidden attack evidence across disparate telemetry sources, achieving a remarkable 2.4x improvement in stage recall for software supply-chain attacks.
LLMs can autonomously fix concurrency bugs, even without prior bug information, by intelligently extracting relevant context using Static Happens-Before Graphs.