1
ATLAS: Agentic Test-time Learning-to-Allocate Scaling
让大模型自主决定测试时计算如何分配，突破固定预算与策略的局限。
arXiv:2606.01667v1 Announce Type: new Abstract: Test-time scaling has become a major way to improve large language model reasoning, but its orchestrat…
让大模型自主决定测试时计算如何分配，突破固定预算与策略的局限。
arXiv:2606.01667v1 Announce Type: new Abstract: Test-time scaling has become a major way to improve large language model reasoning, but its orchestrat…