← Latest papers
🔢 mathematics

On the Slice Rank of Tensors in P-Echelon Form

This paper proves that tensors in PP-echelon form with nonzero diagonal entries possess full slice-rank whenever the Hasse diagram of the underlying poset PP contains no isolated vertices, thereby extending and improving upon recent results by Amanov and Yeliussizov.

Original authors: Omran Ahmadi, Hassan Norouzi

Published 2026-07-22
📖 1 min read🧠 Deep dive

Original authors: Omran Ahmadi, Hassan Norouzi

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Technical Summary: On the Slice Rank of Tensors in P-Echelon Form

Problem Statement
The paper addresses the problem of determining the slice-rank of dd-tensors T:AdFT: A^d \to \mathbb{F}, where AA is a totally ordered finite set and F\mathbb{F} is a field. Specifically, it investigates tensors in PP-echelon form, defined relative to a partially ordered set (poset) P=([d],P)P = ([d], \leq_P). A tensor is in PP-echelon form if, for every tuple (a1,,ad)(a_1, \dots, a_d) in its support, the condition rPsr \leq_P s implies arQasa_r \leq_Q a_s, where Q\leq_Q is the total order on AA.

The central question is whether such tensors, provided they have nonzero diagonal entries (i.e., T(a,,a)0T(a, \dots, a) \neq 0 for all aAa \in A), possess full slice-rank, meaning sr(T)=A\text{sr}(T) = |A|. This generalizes Tao's slice-rank lemma for diagonal tensors. Previous work by Amanov and Yeliussizov established this result for even dd under the stricter condition that the Hasse diagram of PP is connected.

Methodology
The authors employ a functional reformulation of a proposition by Sawin and Tao to establish a lower bound on the slice-rank. The core of the proof relies on Lemma 2.1 (Sawin-Tao), which states that the slice-rank is bounded below by the minimum sum of the sizes of projections of a partition of the set of maximal elements of the tensor's support.

To apply this lemma, the authors introduce a specific ordering strategy in Lemma 2.4. They demonstrate that for any poset PP whose Hasse diagram contains no isolated vertices, one can assign either the original order Q\leq_Q or the reverse order Q\geq_Q to each coordinate j[d]j \in [d]. This assignment ensures that if a diagonal element δa=(a,,a)\delta_a = (a, \dots, a) is less than or equal to a support element xx under the resulting product order, then xx must equal δa\delta_a. This property forces the set of diagonal elements to be contained within the set of maximal elements of the support.

Key Contributions and Results
The paper presents Theorem 1.3, the main result, which asserts:

Let d2d \geq 2. Let AA be a totally ordered finite set, F\mathbb{F} a field, and P=([d],P)P = ([d], \leq_P) a poset whose Hasse diagram has no isolated vertex. If T:AdFT: A^d \to \mathbb{F} is in PP-echelon form with nonzero diagonal entries, then sr(T)=A\text{sr}(T) = |A|.

The proof proceeds by:

  1. Establishing that the diagonal elements Diag(Ad)\text{Diag}(A^d) are a subset of the maximal elements Γ\Gamma of the support under the constructed product order.
  2. Showing that for any partition of Γ\Gamma into dd sets, the sum of the sizes of their coordinate projections is at least A|A|.
  3. Combining this lower bound with the trivial upper bound (achieved by slicing along the first coordinate) to conclude equality.

Significance and Claims
The authors explicitly frame their contribution as an extension and improvement of the results by Amanov and Yeliussizov. The significance is twofold:

  1. Parity Independence: The result holds for both even and odd dd, whereas the previous result by Amanov and Yeliussizov was restricted to even dd.
  2. Relaxed Connectivity Assumption: The paper replaces the requirement that the Hasse diagram of PP be connected with the weaker condition that it has no isolated vertex.

The paper maintains a modest scope, focusing strictly on the theoretical determination of slice-rank for this specific class of tensors. It does not propose new applications, experimental validations, or future implications beyond the mathematical generalization of the existing lemma within the context of the slice-rank method in extremal combinatorics.

Drowning in papers in your field?

Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.

Try Digest →