Search papers, labs, and topics across Lattice.
2
0
4
0
Parallel LLM agents can finally achieve near-linear speedups by strategically parallelizing only when token-based cost estimates predict a net gain, eliminating redundant work and reconciliation overhead.
This work introduces ToolCompass, a post-training framework that guides tool trialing by organizing tool-call representations according to shared functions and jointly reduces intra-function variation across domains and increases inter-function separation.
