Search papers, labs, and topics across Lattice.
Affiliation:
3
0
4
10
Achieving up to 4818 times faster computation in fuzzy PSI protocols could revolutionize privacy-preserving data sharing in real-time applications.
Training speculative decoding models just got an order of magnitude faster, unlocking real-world deployment with a new open-source framework and a suite of production-ready draft models.
PromptTuner slashes SLO violations by up to 7.9x and costs by 4.5x in LLM prompt tuning, outperforming existing resource management systems.
