Search papers, labs, and topics across Lattice.
5
0
6
Tail latency in LLM serving can be cut by up to 50% without relying on length predictions, reshaping how we optimize inference performance.
VIPIR achieves orders-of-magnitude higher throughput for private information retrieval while slashing communication and memory overheads, revolutionizing large-scale database privacy.
Offloading avatar reconstruction can enable over 2.3 times more users in VR while maintaining privacy and minimizing quality loss.
GPIR shatters the PIR performance barrier, achieving 300x speedups on GPUs by rethinking kernel design and data layout to overcome memory bottlenecks exposed by multi-client batching.
Current benchmarks mislead on the security of AI agents against indirect prompt injection; robust defenses require dynamic replanning, constrained LLM-based security checks, and human interaction.
