Bilinear rank of rectangular matrix multiplication

To multiply an a-by-b matrix by a b-by-c matrix, a bilinear algorithm multiplies linear combinations of entries from the two inputs. What is the fewest such products needed for each shape over the rationals?

Why it matters and what counts as progress

Why it matters: Small rectangular products are building blocks for larger algorithms. A lower rank counts fewer scalar multiplications for that shape.

What counts as progress: An upper bound needs an explicit bilinear scheme and an identity valid for every input. Lower bounds rule out all schemes below a stated rank. The number of additions is a separate measure.

How the problem developed

  1. 2026 · Earlier upper bounds [3][4]

    A. I. Perminov and the Lille catalogue Recorded schemes with 644 and 710 products for the two matrix shapes discussed here.

  2. 14 Nov 2026 · Bilinear rank bound [1][2]

    Arhan Canli and Claude agents Gave a 640-product scheme for the 9×109\times 10 by 10×1110\times 11 format and checked the identities in Lean.

  3. 14 Nov 2026 · Bilinear rank bound [5][6]

    Arhan Canli and Claude agents Gave a 708-product scheme for the 9×119\times 11 by 11×1111\times 11 format and checked the identities in Lean.

AI contributions 2

Sources

  1. Quantavane’s rank-640 scheme and provenance
  2. Lean theorem scope and checking method
  3. University of Lille fast matrix multiplication catalogue
  4. Perminov’s rectangular matrix multiplication schemes
  5. Quantavane’s rank-708 scheme and provenance
  6. Lean theorem scope and checking method
Publication history