Symplectic Branching through Crystals

B\'arbara Muniz

REVIEW 3 major objections 4 minor 1 cited by

Symplectic Branching through Crystals

T0 review · 3 major / 4 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read A crystal bijection proves the symplectic branching multiplicity rule.

desk verdict Known theorem, but a genuinely new elementary bijection; the cascade idea is nice, yet the load-bearing Lemma 3.3 is under-specified and needs a careful referee. read the letter →

arxiv 2505.21738 v1 pith:L3ME7BHK submitted 2025-05-27 math.RT math.CO

Bárbara Muniz This is my paper

classification math.RTmath.CO MSC 05E1017B37

keywords symplecticbranchingcrystalsYoungtableauxLittlewood-Richardsonhighestweightrepresentationsquantumgroupssemistandard

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper proves that the multiplicity of a symplectic representation inside a general linear representation—the symplectic branching problem—can be read directly from crystals: it is the number of symplectic-highest-weight semistandard Young tableaux of fixed shape and weight. The proof works by constructing an explicit bijection between the Littlewood–Richardson tableaux that appear in Sundaram's branching rule and these highest-weight tableaux. The bijection is built by alternating two simple deletion maps, one on each side, and is self-contained, avoiding the heavier machinery used in earlier proofs of the same statement. If the construction is correct, the result turns a classical restriction problem into a direct, checkable tableau count.

What carries the argument

The cascade operation: given an ordered sequence $s=(1,2,\dots,m)$ of entries in a tableau, delete the $1$ and replace every remaining $i$ by $i-1$, under hypotheses that preserve semistandardness and the Littlewood–Richardson property. With the two inclusion maps $\iota_{\mathfrak{sp}}$ (delete the last negative-weight entry and slide) and $\iota_{LR}$ (cascade the last feasible sequence, delete the last entry, and slide), the cascade operation generates the recursive bijection $F=\iota_{\mathfrak{sp}}^{-1}\circ F\circ \iota_{LR}$. The load-bearing combinatorial fact is Lemma 3.3, a non-crossing property of consecutive cascade sequences: the relative order of the final entries $m_i$ controls the relative order of the starting $1$s, which is what makes the deletion and sliding choices on the two sides agree at every inductive step.

What would settle it

Search the finite set of tableaux for a semistandard Littlewood–Richardson skew tableau whose cascade sequences satisfy conditions (I)–(III) of Lemma 3.1 but violate the ordering conclusion of Lemma 3.3, for instance $m_i\le m_{i+r}$ while the $1$ of sequence $i$ appears after the $1$ of sequence $i+r$. Running the recursive definition of $F$ on that tableau should then produce a mismatch between the $\iota_{\mathfrak{sp}}$ and $\iota_{LR}$ deletions, yielding two different counts for a small example such as $\lambda=(3,2,1,1)$, $\mu=(2,1)$.

Watch

Extended reading notes

Core claim

The central claim is that for partitions $\lambda$ with at most $2n$ parts and $\mu$ with at most $n$ parts, the multiplicity of $V^{\mathfrak{sp}_{2n}}(\mu)$ in $V^{\mathfrak{gl}_{2n}}(\lambda)$ as a restricted $\mathfrak{sp}_{2n}(\mathbb{C})$-representation equals the number of $\mathfrak{sp}_{2n}$-highest-weight semistandard Young tableaux of shape $\lambda$ and weight $\mu$. The paper establishes this by constructing a bijection $F$ between Sundaram's symplectic Littlewood–Richardson tableaux of shape $\lambda\setminus\mu$ and weight $(2\delta)'$ and the $\mathfrak{sp}_{2n}$-highest-weight tableaux of shape $\lambda$ and weight $\mu$. The bijection is recursive: each step removes one ordered $\{i,\bar{i}\}$-pair from the highest-weight side and one cascaded sequence from the Littlewood–Richardson side, and the two operations are inverse to each other. The proof first handles the stable range $\ell(\lambda)\le n$, then identifies the $n$-symplectic condition that selects the correct tableaux when $n$ is not large enough, and finally invokes Sundaram's branching rule to equate the two counts.

Load-bearing premise

Everything rests on a combinatorial ordering property of the cascade operation: sequences of consecutive fillings must obey a non-crossing rule (Lemma 3.3), and if that rule failed for even one tableau the two sides of the induction could delete different boxes and the bijection would break.

Editorial extensions

If this is right

Symplectic branching multiplicities can be computed by counting highest-weight tableaux, so restriction coefficients become directly visible in the crystal graph.
The bijection is explicit and recursive, so a concrete multiplicity can be computed by iterating delete, cascade, and slide steps without invoking the Robinson–Schensted–Knuth correspondence or other external machinery.
In the stable range $\ell(\lambda)\le n$, the restriction multiplicity is independent of $n$, and the bijection is a pure translation between type C and type A tableau conditions.
For $\ell(\lambda)>n$, exactly the $n$-symplectic Littlewood–Richardson tableaux survive, identifying the precise boundary where small $n$ cuts off the stable rule.
The main theorem confirms that the crystal model and Sundaram's branching rule describe the same multiplicities, closing the original conjecture through a self-contained combinatorial proof.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

The cascade-and-slide recursion looks like a tableau insertion algorithm; it may generalize to other classical branchings, such as orthogonal restrictions, by replacing the $\{i,\bar{i}\}$-pair decomposition with the analogous signed alphabet.
The $n$-symplectic condition introduced here is a tableau-level stability cutoff; one could test whether it matches known stabilization thresholds for other symplectic or orthogonal branching rules.
A direct computational implementation of $F$ for random $\lambda,\mu$ would stress-test Lemma 3.3; any violation would show up as a mismatch with Sundaram's tableaux, making the lemma's role falsifiable.
The proof identifies the non-crossing property of cascades as the core combinatorial content; weakening that property might yield new branching rules or generalized tableau models for other quantum symmetric pairs.

Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Let me give you my read on Muniz's paper. The theorem itself is not new—Schumann–Torres and Naito–Suzuki–Watanabe already proved the Naito–Sagaki branching rule. What is new is the proof mechanism: a self-contained bijection between sp-highest-weight tableaux and Sundaram's Littlewood–Richardson tableaux, built from a simple cascading operation on sequences of {i, ī}-pairs. I think the cascade operation is a good idea, and the paper is honest about what it is doing; the construction is concrete and mostly checkable by hand.

The soft spots are concentrated where the reader's report points. Lemma 3.3 is load-bearing: it is exactly what makes the inductive bijection well-defined, and it is not stated or proved cleanly. The statement fixes r0 < r and concludes something for i < r0, but the application in Proposition 4.3 needs r0 = 1 and r0 = r, which are edge cases the proof doesn't explicitly cover. The proof also hand-waves the claim that interference between cascades happens at most once per box, and the induction is written as if i were the induction variable when it is the cascade index. In Proposition 4.3 there is the line 'i1 ≤ i2 if and only if i1 ≥ i2', which has to be a typo, but it sits right at the point where the comparison is being made, so it is genuinely confusing rather than cosmetic. These are exactly the places where a subtle sign error could break the bijection.

On the other side, I did not find a clear fatal error. The overall structure is plausible: F is constructed independently of the asserted formula, Sundaram's rule is used only as an external final step, and there is no circularity. Lemma 4.5 is abbreviated but the inequalities there look like they can be derived with a bit more care; it is a minor concern relative to Lemma 3.3.

So my verdict: the paper deserves a serious referee, not a desk reject. The referee should be asked to check Lemma 3.3 and Proposition 4.3 line by line, and I would probably insist on a small exhaustive verification for small shapes before publication. The typos are numerous enough that the paper needs a careful revision regardless.

Who is this for? People working on classical branching rules and crystal bases. The cascade operation might be reusable for other stable-to-unstable branchings, and the paper makes the connection between Sundaram tableaux and highest-weight crystals much more transparent than the earlier proofs. I wouldn't cite it for the theorem, but I would cite it for the bijection if it survives referee scrutiny.

Recommendation: engage with it, send it out, but require the lemma fix and a clean rewrite before acceptance.

Referee Report

3 major / 4 minor

Summary. The paper proposes an alternative proof of the Naito–Sagaki conjecture on symplectic branching for the restriction of gl_{2n}(C)-representations to sp_{2n}(C). The strategy is to construct an explicit bijection F_{λ,μ} between Littlewood–Richardson tableaux of skew shape λ\μ and weight (2δ)' and sp_{2n}-highest weight semistandard Young tableaux of shape λ and weight μ. The bijection is defined recursively using two deletion/sliding maps ι_sp and ι_LR, and the main technical work is to show that the recursion is well-defined and lands in the correct domain. In the finite-n case, the n-symplectic condition is introduced to select those tableaux whose images have entries in the alphabet A_n. The main theorem then follows by combining the bijection with Sundaram's branching rule. The result itself is already known, so the paper's contribution is an alternative, putatively elementary and self-contained proof.

Significance. If the proof is completed, this would provide a genuinely elementary and self-contained bijective proof of a conjecture previously proved via Burge correspondence and symplectic RSK or via quantum symmetric pairs. The construction cleanly separates the stable case from the finite-n case and identifies a simple n-symplectic condition, which is conceptually appealing. The manuscript does not include machine-checked code or formal proofs, and the paper's main novelty—the bijection—is precisely where the current proof has gaps. Since the underlying theorem is known to be true, those gaps are plausibly fixable, but as written the proof is not yet convincing.

major comments (3)

[§3, Lemma 3.3] Lemma 3.3 is stated with conflicting quantifiers: it fixes r0 < r, assumes m_i > m_{i+r} for all i ≤ r0, and concludes a statement about k ∈ s_i and k ∈ s_{i+r} only for i < r0. The conclusion is therefore vacuous for i = r0, while Proposition 4.3 needs exactly the case i = 1 when r0 = 1 and the case i = r when r0 = r. The proof also introduces the claim that interference between cascades 'happens at most once per box' without proof, and the phrase 'by induction on i' does not identify the induction variable of the lemma. Since Proposition 4.3 reduces the well-definedness of F to precisely this lemma, the current version does not establish the main bijection.
[§4, Proposition 4.3] The sentence 'We then have i1 ≤ i2 if and only if i1 ≥ i2 if and only if 1∈ s1 is below 1 ∈ s2' is not a meaningful equivalence; at minimum it contains a typo that reverses one of the inequalities. The intended statement must relate the row indices i_k of the 1's in the cascaded sequences to the lengths m_k, and this is the only argument connecting the ι_sp side to the ι_LR side of the induction. As written, the proof of F(Im ι_LR) = Im ι_sp is incomplete.
[§4.1, Lemma 4.5] The proof of Lemma 4.5 is too compressed for a step that is needed to pass from the stable bijection to Sundaram's rule. In the (⇐) direction, the sentence 'That gives us an entry Y(n+i',j')=2m ≤ 2(m+i) ≤ 2i'' does not follow from the preceding claim without specifying the row i' and how the bound 2i' is obtained; in the (⇒) direction, the final inequality '2(i−1) < ι_LR(Y)(n+i−1,1) ≤ m−2' is asserted without derivation. Because Lemma 4.5 identifies the domain of F with the n-symplectic LR tableaux counted by Sundaram's rule, this gap affects the main theorem in the general case.

minor comments (4)

[§2, Lemma 2.2] The proof of Lemma 2.2 says the case φ_i(v′) < ε_i(a_N) 'is absurd, as it implies φ_i(v′) < 0'; this is correct only because e_i(a_N) = 0 forces ε_i(a_N) = 0. The sentence would be clearer if that point were stated explicitly.
[§4, Definition 4.3] The phrase 'anti-lexicographic order on the reading' is used to define the 'last ordered sequence' s, but the order is not defined explicitly; the proof of Lemma 4.2 depends on this choice and would benefit from a precise definition.
[Examples 4.1 and 4.3] The overlines that distinguish negative entries in the alphabet A_n are not visible in several displays, and the tableaux are not aligned with the arrows; this makes it hard to verify the effect of ι_sp and ι_LR on the examples.
[Introduction] The reference [KT24] is cited as 'Kumar–Tores' in the text; the second author's name is Torres. Minor spelling inconsistencies should be corrected throughout.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the branching multiplicity is matched to Sundaram's external rule via an independently constructed bijection.

full rationale

The derivation is not circular. The multiplicities on the left-hand side are taken from Sundaram's branching rule [Sun90], an external theorem independent of the paper's construction. The bijection F is defined recursively in Definition 4.4 and is shown in Proposition 4.3 to map Littlewood-Richardson tableaux of shape lambda\mu and weight (2 delta)' to sp-highest weight tableaux of shape lambda and weight mu; neither side of this map is used as an input to the other, and the proof of bijectivity relies on the explicit deletion and cascading maps iota_sp, iota_LR and on Lemma 3.3 rather than on the multiplicity formula. Lemma 4.5 then identifies the n-symplectic condition as exactly the condition under which F's outputs lie in the finite alphabet A_n, so the final equality LHS = sum_delta (sp nc)^lambda_{mu,(2 delta)'} = |Im F intersect B_{gl}(lambda)| = RHS is a genuine comparison of two independently defined sets. There are no fitted parameters, no self-citations, and no uniqueness claim or ansatz imported from the author's prior work. The apparent difficulties in Lemma 3.3 and the typo in Proposition 4.3 concern the correctness of the proof, not circularity.

Assumptions & free parameters 0 free parameters · 5 assumptions · 0 invented entities

The proof relies on standard crystal theory (existence, tensor rule, tableau model) and on Sundaram's external branching theorem. No free parameters or invented entities are introduced. The non-trivial choice of embedding into tensor powers is flagged in Remark 2.1.

assumptions (5)

standard math Existence and uniqueness of crystal bases for finite-dimensional U_q(g)-modules (Theorem 1.3).
Used in Section 1.3 to justify the crystal model, cited from Hong-Kang.
standard math Tensor product rule for crystals (Theorem 1.4).
Used to characterize sp-highest weight vectors in Lemma 2.2.
standard math The crystal of a U_q(gl_n) irreducible module is the set of semistandard Young tableaux of shape λ on alphabet [n] (Proposition 2.1).
The tableau model used throughout; cited from Hong-Kang.
domain assumption Sundaram's symplectic branching rule (stated as a theorem in the Introduction).
External theorem used at the end of the Main Theorem proof to equate the multiplicity with the number of n-symplectic Littlewood-Richardson tableaux; not reproved.
domain assumption The symplectic crystal structure comes from a fixed embedding of V(λ) into tensor powers of the natural representation (Remark 2.1).
The paper notes the conjecture is false for other embeddings, so this choice is load-bearing.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Symplectic Branching through Crystals." pith.science (2026). https://pith.science/paper/L3ME7BHK

@misc{pith2026250521738,
  author       = {Pith},
  title        = {Pith review of: Symplectic Branching through Crystals},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/L3ME7BHK}},
  note         = {Machine review of arXiv:2505.21738}
}

abstract

We give an alternative proof of Naito--Sagaki's conjecture, which states that the restriction of $gl_{2n}(\mathbb{C})$-representations to $sp_{2n}(\mathbb{C})$ can be described in terms of crystals. Using the tableau model for crystals, we construct an explicit and self-contained bijection between their highest weight elements and Sundaram's branching model.

Discussion (0). Sign in to comment.

Forward citations

Cited by 1 Pith paper

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

The recording tableaux in the quantum Littlewood-Richardson map, the orthogonal transpose symmetry map, and the computation of $\mathfrak{k}$-highest weight tableaux
math.CO 2026-03 unverdicted novelty 6.0 of 10

Combinatorial surjectivity of Watanabe’s quantum LR map is proved via reverse Schensted insertion and inverse reduction, restricting orthogonal transpose symmetry and characterizing 𝔨-highest weight tableaux by inequalities.

Reference graph

Works this paper leans on

16 extracted references · 15 canonical work pages · cited by 1 Pith paper

[1]

Four correspondences between graphs and generalized Y oung tableaux

William H Burge. Four correspondences between graphs and generalized Y oung tableaux. Journal of Combinatorial Theory, Series A , 17(1):12--30, 1974

work page 1974
[2]

Introduction to quantum groups and crystal bases

Jin Hong and Seok-Jin Kang. Introduction to quantum groups and crystal bases . American Mathematical Society, 2002

work page 2002
[3]

Lectures on quantum groups

Jens Carsten Jantzen. Lectures on quantum groups . American Mathematical Society, 2009

work page 2009
[4]

Quantum symmetric K ac-- M oody pairs

Stefan Kolb. Quantum symmetric K ac-- M oody pairs. Advances in Mathematics , 267:395–469, December 2014

work page 2014
[5]

The honeycomb model of gln(c) tensor products i: Proof of the saturation conjecture

Allen Knutson and Terence Tao. The honeycomb model of gln(c) tensor products i: Proof of the saturation conjecture. Journal of the American Mathematical Society , 12(4):1055--1090, 1999

work page 1999
[6]

The branching models of Kwon and Sundaram via flagged hives

V. Sathish Kumar and Jacinta Torres. The branching models of K won and S undaram via flagged hives. arXiv:2412.19721 , 2024

work page Pith review arXiv 2024
[7]

Combinatorial extension of stable branching rules for classical groups

Jae-Hoon Kwon. Combinatorial extension of stable branching rules for classical groups. Transactions of the American Mathematical Society , 370(9):6125--6152, 2018

work page 2018
[8]

The theory of group characters and matrix representations of groups

Dudley Littlewood. The theory of group characters and matrix representations of groups . Clarendon Press, Oxford, 1940

work page 1940

Show all 16 references

[9]

Paths and root operators in representation theory

Peter Littelmann. Paths and root operators in representation theory. Annals of Mathematics , 142(3):499--525, 1995

1995
[10]

Combinatorics of generalized exponents

C \'e dric Lecouvey and Cristian Lenart. Combinatorics of generalized exponents. International Mathematics Research Notices , 2020(16):4942--4992, 2020

2020
[11]

An approach to the branching rule from sl_ 2n ( C ) to sp_ 2n ( C ) via L ittelmann's path model

Satoshi Naito and Daisuke Sagaki. An approach to the branching rule from sl_ 2n ( C ) to sp_ 2n ( C ) via L ittelmann's path model. Journal of Algebra , 286(1):187--212, 2005

2005
[12]

A proof of the N aito-- S agaki conjecture via the branching rule for quantum groups

Satoshi Naito, Yujin Suzuki, and Hideya Watanabe. A proof of the N aito-- S agaki conjecture via the branching rule for quantum groups. arXiv:2502.07270 , 2025

2025
[13]

A non- L evi branching rule in terms of L ittelmann paths

Bea Schumann and Jacinta Torres. A non- L evi branching rule in terms of L ittelmann paths. Proceedings of the London Mathematical Society , 117(5):1077--1100, 2018

2018
[14]

On the combinatorics of representations of the symplectic group

Sheila Sundaram. On the combinatorics of representations of the symplectic group. Thesis (Ph. D.)--Massachusetts Institute of Technology , 1986

1986
[15]

Tableaux in the representation theory of the classical L ie groups

Sheila Sundaram. Tableaux in the representation theory of the classical L ie groups. Institute for Mathematics and Its Applications , 19, 1990

1990
[16]

Symplectic tableaux and quantum symmetric pairs

Hideya Watanabe. Symplectic tableaux and quantum symmetric pairs. arXiv:2308.01718 , 2023

2023 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Four correspondences between graphs and generalized Y oung tableaux

William H Burge. Four correspondences between graphs and generalized Y oung tableaux. Journal of Combinatorial Theory, Series A , 17(1):12--30, 1974

work page 1974

[2] [2]

Introduction to quantum groups and crystal bases

Jin Hong and Seok-Jin Kang. Introduction to quantum groups and crystal bases . American Mathematical Society, 2002

work page 2002

[3] [3]

Lectures on quantum groups

Jens Carsten Jantzen. Lectures on quantum groups . American Mathematical Society, 2009

work page 2009

[4] [4]

Quantum symmetric K ac-- M oody pairs

Stefan Kolb. Quantum symmetric K ac-- M oody pairs. Advances in Mathematics , 267:395–469, December 2014

work page 2014

[5] [5]

The honeycomb model of gln(c) tensor products i: Proof of the saturation conjecture

Allen Knutson and Terence Tao. The honeycomb model of gln(c) tensor products i: Proof of the saturation conjecture. Journal of the American Mathematical Society , 12(4):1055--1090, 1999

work page 1999

[6] [6]

The branching models of Kwon and Sundaram via flagged hives

V. Sathish Kumar and Jacinta Torres. The branching models of K won and S undaram via flagged hives. arXiv:2412.19721 , 2024

work page Pith review arXiv 2024

[7] [7]

Combinatorial extension of stable branching rules for classical groups

Jae-Hoon Kwon. Combinatorial extension of stable branching rules for classical groups. Transactions of the American Mathematical Society , 370(9):6125--6152, 2018

work page 2018

[8] [8]

The theory of group characters and matrix representations of groups

Dudley Littlewood. The theory of group characters and matrix representations of groups . Clarendon Press, Oxford, 1940

work page 1940