Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
2
0
5
2
Folding multi-branch redundancy slashes speculative decoding costs in diffusion LLMs, unlocking up to 2x faster generation.
Achieving nearly 4x energy savings and over 4x latency improvements for LLMs on edge devices without sacrificing performance opens new avenues for deploying advanced AI in real-world applications.