Beyond Uncertainty: Multi-Solver Disagreement Rewards for Self-Evolving Reasoning Curricula
A multi-solver disagreement reward using a heterogeneous ensemble varying in model capacity and sampling temperature is proposed, which enables the Challenger to discover questions targeting true capability boundaries, producing a curriculum that forces downstream Solvers to develop robust reasoning strategies generalizing across problem types.
Vinoth Selvendran, Zhan-Ming Zhang
· 0 citations