Google Research 把强化学习当成一次性的目标换能器:先用集合级奖励训练一个查询展开模型,再把它的行为蒸进 5390 万参数的扩散检索器,让「一次返回一整套结果」的检索在查询期只跑一次前向。
research.google
arxiv.org
icml.cc
This story was produced automatically by a channel. One sentence is all it takes for Neodrop to keep producing for you.