Recovering an LCS in O(n2/w) Time and Space

Here we make use of word-level parallelism to recover a longest common subsequence of two input strings both of length n in O(n 2 /w)  time and space, where w is the number of bits in a machine word. For the special case where one of the input atrings is close to w its complexity is reduced to linear time and space. Keywords: Longest Common Subsequence, Bit-parallelism.