Given array of length n. I need to find amount of subarrays (or amount of pairs 1 <= l <= r <= n) such that OR(a[l], a[l+1], ..., a[r]) > max(a[l], a[l+1], ..., a[r]) for O(nlogn * T), where T = max(a[i]).

My approach: we can build Sparse Table for min and bitwise OR to get result on the segment for O(1). Also I noticed that OR(a[l], a[l+1], ..., a[r]) >= max(a[l], a[l+1], ..., a[r]), and OR(a[l], a[l+1], ..., a[r]) <= 2^(logT + 1) - 1, and both functions are monotonically increasing when we increment R. Not sure how to do the rest…