归一化后加权
归一化后加权
def normalize(scores):
lo, hi = min(scores), max(scores)
span = (hi - lo) or 1.0
return [(s - lo) / span for s in scores]
fused = [
alpha * s_dense + (1 - alpha) * s_sparse
for s_dense, s_sparse in zip(normalize(dense_scores), normalize(sparse_scores))
]归一化与权重都要在带标注的查询集上调：先固定 top-k 候选池，再网格搜索 alpha 这类权重。