Search papers, labs, and topics across Lattice.
Case Western Reserve University
4
0
7
Hybrid-thinking LLMs can be dramatically improved by simply separating the feed-forward pathways for reasoning and non-reasoning modes, leading to less leakage and better accuracy.
Quantum cloud providers can now efficiently schedule parallel jobs on modular QPUs, optimizing qubit mapping and teleportation for fair resource allocation.
OpenQASM 3.0's dynamic quantum circuits can now be efficiently executed on CUDA-Q, unlocking performance gains by avoiding branch duplication and leveraging low-latency classical feedback.
Agent evaluation is bottlenecked by environment interaction overhead, but ACE-Bench slashes this by using static JSON files, enabling fast and reproducible training-time validation.
