A 58-Addition, Rank-23 Scheme for General 3x3 Matrix Multiplication
Abstract
This paper presents a new state-of-the-art algorithm for exact 3times3 matrix multiplication over general non-commutative rings, achieving a rank-23 scheme with only 58 scalar additions. This improves the previous best additive complexity of 60 additions without a change of basis. The result was discovered through an automated search combining ternary-restricted flip-graph exploration with greedy intersection reduction for common subexpression elimination. The resulting scheme uses only coefficients from {-1, 0, 1}, ensuring both efficiency and portability across arbitrary fields. The total scalar operation count is reduced from 83 to 81.
Community
A 58-addition, rank-23 scheme for exact 3×3 matrix multiplication sets a new SOTA. This improves the previous best of 60 additions without basis change. The scheme uses only ternary coefficients {-1,0,1} and was discovered via combinatorial flip-graph search + greedy CSE heuristics.
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Parallel Heuristic Exploration for Additive Complexity Reduction in Fast Matrix Multiplication (2025)
- Fast Matrix Multiplication via Ternary Meta Flip Graphs (2025)
- Faster Algorithms for Structured Matrix Multiplication via Flip Graph Search (2025)
- Low-Rank GEMM: Efficient Matrix Multiplication via Low-Rank Approximation with FP8 Acceleration (2025)
- Graded Projection Recursion (GPR): A Framework for Controlling Bit-Complexity of Algebraic Packing (2025)
- Utilizing Circulant Structure to Optimize the Implementations of Linear Layers (2025)
- Emulation of Complex Matrix Multiplication based on the Chinese Remainder Theorem (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper