CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks

By Fanzhe Meng · Paper · cs.LG

Training terminal agents requires executable and verifiable tasks that are not merely solvable, but appropriately challenging for learning. Executable validation establishes feasibility, yet does not reveal how a task behaves relative to a given solver setting. In this paper, we

Cs.lg

View original

HomeResourceLoading…